Banana root system: towards a better understandingfor its ...

Banana Root System: towards a better understandingfor its productive managementProceedings of an international symposium held in San José, Costa Rica, 3-5 November 2003

Sistema Radical del Banano: hacia un mejor conocimiento para su manejo productivoMemorias de un simposio internacional, San José, Costa Rica, 3-5 noviembre 2003

David W. Turner and Franklin E. Rosales, editors

INIBAP – International Network for the Improvement of Banana and Plantain The mission of INIBAP is to sustainably increase the productivity of banana and plantain grown on smallholdings for domestic consumption and for local and export markets.The Programme has four specific objectives:• To organize and coordinate a global research effort on banana and plantain, aimed at the

development, evaluation and dissemination of improved cultivars and at the conservation and use of Musa diversity

• To promote and strengthen collaboration and partnerships in banana-related research activities at the national, regional and global levels

• To strengthen the ability of NARS to conduct research and development activities on bananas and plantains

• To coordinate, facilitate and support the production, collection and exchange of information and documentation related to banana and plantain.

INIBAP is a network of the International Plant Genetic Resources Institute (IPGRI), a Future Harvest centre.

MUSALAC – Banana and Plantain Research and Development Network for Latin America and the CaribbeanMUSALAC was created under the umbrella of FORAGRO on 6 June 2000 in Cartagena de Indias, Colombia, following the signing of a Constitution Agreement. MUSALAC is composed of 15 national research and development institutions representing their respective country (Bolivia, Brazil, Colombia, Costa Rica, Cuba, Ecuador, Honduras, Jamaica, Mexico, Nicaragua, Panama, Peru, Puerto Rico, Dominican Republic and Venezuela) and 4 regional/international institutions (CATIE, CIRAD, IICA and INIBAP). The main objective of MUSALAC is to increase the productivity and competitiveness of the plantain and banana commodity chain through scientific and technological development by strengthening national research and development programmes, facilitating exchanges between stakeholders and prioritizing and coordinating actions in Latin America and the Caribbean. MUSALAC is led by a Steering Committee composed of one representative from each member country; one President and two Vice-presidents and an Executive Coordinator, which is INIBAP-LAC, headquartered at CATIE in Turrialba, Costa Rica.

CORBANA – National Banana Corporation S.A.CORBANA S.A. is a public, non-profit institution, with the following objectives:• To strengthen research on the banana crop• To increase banana productivity at a minimum environmental risk.• To promote programmes on cost reduction• To provide services in research, technical assistance and information on prices and markets.• To facilitate an equal relationship between national producers and export companies.• To establish jointly with the Costa Rican Government, banana policies to help maintaining the industry

on the long term.• To centralize banana information to promote and facilitate the participation of the banana sector in

research and technological development of the banana sector.

Citation: Turner D.W. and F.E. Rosales (eds). 2005. Banana Root System: towards a better understanding for its productive management: Proceedings of an international symposium / Sistema Radical del Banano: hacia un mejor conocimiento para su manejo productivo: Memorias de un simposio internacional. International Network for the Improvement of Banana and Plantain, Montpellier, France.

Cover illustration: Transverse section of a mature root of ‘Williams’ (Musa AAA, Cavendish) showing aerenchyma development in the mid cortex. (Credit: Michael W. Shane).INIBAP ISBN: 2-910810-61-5 © International Plant Genetic Resources Institute 2005

INIBAPParc Scientifique Agropolis II34397 Montpellier Cedex 5France

MUSALACC/o CATIEApartado 607170 Turrialba, Costa Rica

CORBANA S.A.A.A. 6504-100 San JoséCosta Rica

Banana Root System: towards a better understandingfor its productive management

Proceedings of an international symposium held in San José, Costa Rica, 3-5 November 2003

Sistema Radical del Banano: hacia un mejor conocimiento

para su manejo productivo

Memorias de un simposio internacional, San José, Costa Rica, 3-5 noviembre 2003

David W. Turner and Franklin E. Rosales, editors

The problem of banana root deterioration and its impact on production2 C.A. Gauggel et al. 3

AcknowledgementsThe International Network for the Impro-vement of Banana and Plantain (INIBAP) and the National Banana Corporation (CORBANA, S.A.), organizers of the International Symposium “Banana root system: towards a better understanding for its productive management”, underta-ken by INIBAP and CORBANA, wish to thank the sponsors and co-sponsors who, with their contribution, made possible the development of such an important event for the benefit of the banana sector worldwide.

ReconocimientosLa Red Internacional para el Mejora-miento del Banano y el Plátano (INIBAP) y la Corporación Bananera Nacional (CORBANA, S.A.), organizadores del Symposium Internacional “Sistema radical del banano: hacia un mejor cono-cimiento para su manejo productivo”, agradecen a los patrocinadores y cola-boradores quienes, con su contribución, hicieron posible el desarrollo de este evento tan importante para el beneficio del sector bananero mundial.

Sponsors/Patrocinadores:• Lefruit, Ecuador

• Dole Fresh Fruit International & Standard Fruit Co. of Costa Rica S.A.

• Abonos Del Pacífico S.A. (ABOPAC), Costa Rica

• Bayer CropScience, Costa Rica

Co-sponsors/Co-patrocinadores:• Grupo Colono División Agropecuaria, Costa Rica

• BASF of Costa Rica

• Makhteshim-Agan, Costa Rica

• Grupo Trisan S.A. and Kemira Grow How, Costa Rica

• Industrias Bioquím Centroamericana S.A., Costa Rica

• Syngenta, Costa Rica

• Chiquita Brands


Contentsi Preface

ii Objective

1. Banana root deterioration and impacts on production1.1 The problem of banana root deterioration and its impact on production: Latin America’s experience

C.A. Gauggel, F. Sierra and G. Arévalo 13

1.2 Research in progress and future perspectives on root system management (Abstract)

D.H. Marín 23

1.3 Banana yield response to nematode control in Dole farms in Costa Rica (Abstract)

M. Castro, R. Duarte, C. Portillo and J. Gonzales 24

1.4 Relationship between functional root content and banana yield in Costa Rica

E. Serrano 25

2. Root anatomy and morphology2.1 Anatomy and morphology of monocotyledonous and dicotyledonous roots

N. Vásquez M. 37

2.2 Methodologies for root system assessment in bananas and plantains (Musa spp.)

G. Blomme, K. Teugels, I. Blanckaert, G. Sebuwufu, R. Swennen and A. Tenkouano 43

2.3 Distribution of banana roots in time and space: new tools for an old science

X. Draye, F. Lecompte and L. Pagès 58

2.4 Development and formation of plantain roots (Musa AAB Simmonds)

S. Belalcázar C. F.E. Rosales and L.E. Pocasangre 75

2.5 Stratification and spatial distribution of the banana (Musa AAA, Cavendish subgroup, cvs ‘Valery’ and ‘Grande Naine’) root system

M. Araya 83 3. Root physiology 3.1 Factors affecting the physiology of the banana root system

D.W. Turner 107

3.2 Ion absorption and proton extrusion by banana roots

B. Delvaux and G. Rufuikiri 114


4. Soils and root development 4.1 Soil physical properties and banana root growth

R. Vaquero M. 125

4.2 Interrelations between the soil chemical properties and the banana plant root system

C.A. Gauggel, D. Moran and E. Gurdian 132

4.3 Banana soil acidification in the Caribbean coast of Costa Rica and its relationship with increased aluminium concentrations

E. Serrano 142

4.4 Banana root and soil health (BRASH) project – Australia

T. Pattison, L. Smith, P. Moody, J. Armour, K. Badcock, J. Cobon, V. Rasiah, S. Lindsay and L. Gulino 149

5. Pathogen – root system interactions 5.1 The potential use of microbial communities inside suppressive banana plants for banana root protection

A. zum Felde, L. Pocasangre and R.A. Sikora 169

5.2 Effect of arbuscular mycorrhizal fungi (AMF) and other rhizosphere micro-organisms on development of the banana root system

M.C. Jaizme-Vega, A.S. Rodríguez-Romero and M.S. Piñero Guerra 178

5.3 Biological control of nematodes in banana

E. Fernández, J. Mena, J. González and M.E. Márquez 193

5.4 Parasitic nematodes on Musa AAA (Cavendish subgroup cvs ‘Grande Naine’, ‘Valery’ and ‘Williams’)

M. Araya and T. Moens 201

5.5 The effect of arbuscular mycorrhizal fungi (AMF) - nematode interactions on the root development of different Musa genotypes

A. Elsen, R. Swennen and D. De Waele 224

5.6 Secondary metabolites in roots and implications for nematode resistance in banana (Musa spp.)

N. Wuyts, G. Lognay, L. Sági, D. De Waele and R. Swennen 238

6. Recommendations 2477. List of participants 251


Contenidoi Prefacio

ii Objetivo

1. Deterioro radical del banano y sus impactos en la producción1.1 La problemática del deterioro radical del banano y su impacto en la producción: Experiencia en América Latina

C.A. Gauggel, F. Sierra y G. Arévalo 13

1.2 Investigaciones en progreso y perspectivas futuras para el manejo del sistema radical (Resumen)

D.H. Marín 23

1.3 Respuesta en producción de banano al control de nematodos en fincas de Dole en Costa Rica (Resumen)

M. Castro, R. Duarte, C. Portillo y J. Gonzáles 24

1.4 Relación entre el contenido de raíz funcional y la producción de banano en Costa Rica

E. Serrano 25

2. Anatomía y morfología de la raíz 2.1 Anatomía y morfología de raíces monocotiledóneas y dicotiledóneas

N. Vásquez M. 37

2.2 Metodologías para evaluar el sistema radical en bananos y plátanos (Musa spp.)

G. Blomme, K. Teugels, I. Blanckaert, G.Sebuwufu, R. Swennen y A. Tenkouano 43

2.3 Distribución de las raíces de banano en tiempo y espacio: nuevas herramientas para una ciencia antigua

X. Draye, F. Lecompte y L. Pagés 58

2.4 Desarrollo y formación de las raíces de plátano (Musa AAB Simmonds)

S. Belalcázar C. F.E. Rosales y L.E. Pocasangre 75

2.5 Estratificación y distribución espacial del sistema radical del banano (Musa AAA, subgrupo Cavendish, cvs ‘Valery’ y ‘Grande naine’)

M. Araya 83

3. Fisiología de la raíz 3.1 Factores que afectan la fisiología del sistema radical del banano

D.W. Turner 107


3.2 Absorción de iones y extrusión de protones por las raíces de banano

B. Delvaux y G. Rufuikiri 114

4. Suelos y desarrollo radical 4.1 Propiedades físicas del suelo y desarrollo del sistema radical

R. Vaquero M. 125

4.2 Interrelaciones entre las propiedades químicas del suelo y el sistema radical del banano

C.A. Gauggel, D. Moran y E. Gurdián 132

4.3 Acidificación de los suelos bananeros en la región Caribe de Costa Rica y su relación con el incremento en las concentraciones de aluminio

E. Serrano y A. Segura 142

4.4 Proyecto sobre salud del suelo y raíces de banano - Australia

T. Pattison, L. Smith, P. Moody, J. Armour, K. Badcock, J. Cobon, V. Rasiah, S. Lindsay y L. Gulino 149

5. Interacciones patógeno – sistema radical 5.1 Uso potencial de las comunidades microbianas en plantas supresivas de banano para proteger las raíces del banano

A. zum Felde, L.E. Pocasangre y R.A. Sikora 169

5.2 Efecto de los hongos micorrícicos arbusculares (HMA) y de otros microorganismos de la rizosfera en el desarrollo del sistema radical del banano

M.C. Jaizme-Vega, A.S. Rodríguez-Romero y M.S. Piñero Guerra 178

5.3 Control biológico de nematodos en banano

E. Fernández, J. Mena, J. González y M.E. Márquez 193

5.4 Nematodos parásitos de Musa AAA (subgrupo Cavendish cvs ‘Grande naine’, ‘Valery’ y ‘Williams’)

M. Araya y T. Moens 201

5.5 Efecto de las interacciones entre hongos micorrícicos arbusculares (HMA) y nematodos en el desarrollo radical de diferentes genotipos de Musa

A. Elsen, R. Swennen y D. De Waele 224

5.6 Metabolitos secundarios en raíces y sus implicaciones para la resistencia a nematodos en banano (Musa spp.)

N. Wuyts, G. Lognay, L. Sági, D. De Waele y R. Swennen 238

6. Recomendaciones 247

7. Lista de participantes 251


PrefaceThe fruit of bananas and plantains (Musa spp.) being delicious and the aerial parts of the plant spectacular, it is easy to overlook the root system that supports them. Many scientists have commented on our limited knowledge of banana and plantain roots. However, those who work on root systems, especially in plantations, are aware that a huge effort is needed to extract even a small piece of information.

This volume contains papers presented during the International symposium on the ‘Banana Root System: towards a better understanding for its productive management’ held in San José, Costa Rica, in November 2003. Although root health is a universal preoccupation, the idea to hold a symposium on the topic follows from concerns expressed by the banana producing/exporting sector, especially Romano Orlich, President of the National Banana Corporation from Costa Rica. CORBANA asked INIBAP to work with them in exploring solutions to the problems of the banana root system. This Symposium shows that the private and public sector can share a research agenda.

The objective of the symposium was to present the state of the art knowledge on the functioning of the Musa root system and its interrelationships with micro-organisms and pests, and to propose ways to improve root health. The demand for this kind of information is increasing because of the need to prevent deterioration of root health, which contributes to a steady decline in production in many commercial farms of Latin American and the rest of the world. The papers present work done on cells as well as whole plants, and short-term studies as well as ones that cover many years and multiple locations. An appreciation of these differences in scale is important because knowledge gained about root tissues in short-term experiments, for example, needs to be integrated with our knowledge at higher levels of plant organization and over longer time scales. Similarly, laboratory results on root structure and function, and on roots and microbes, need to be compared with results from field experiments. Only by facilitating interactions between people working at different scales, will we increase our understanding of banana and plantain root systems, in the laboratory as well as in the field. This is exciting in its own right, and provides the basis for sound management decisions needed to sustain our markets and our environment.

Work on banana and plantain root systems is usually done within the broader international framework of research on root systems in plants. This is a valuable approach. Good progress is being made on evaluating the impact of the pests that attack the root system. We are learning about the importance of having a soil that is amenable to management, such as drainage, irrigation and fertilization. We should not push this knowledge aside as we explore other aspects of root performance. On the other hand, we are making slow progress in understanding the role of genetics in the functioning and architecture of the banana root system. Indeed, we have yet to ask the question: what sort of root system do we want for banana plants? The answer may be somewhat different, according to the local situation, but there will be some general principles as the plants will always need water, nutrients and being able to stay upright! These answers may help us see clearly what is important and help us focus on how best to achieve our objectives. In the final session of the symposium, the participants produced an outline on the way ahead. These points reflect a consensus following the diversity of ideas and information presented during the symposium.

We would like to thank the contributing authors for their insights into the problem, and INIBAP and CORBANA for bringing us together and making this publication available worldwide. We hope that these proceedings will stimulate readers into exploring how this knowledge might be translated into improved management practices.

David W. TurnerThe University of Western Australia

Perth, Australia

Franklin E. RosalesINIBAP Latin America and the Caribbean regional office

Turrialba, Costa Rica


PrefacioEl fruto de los bananos y plátanos (Musa spp.) es delicioso y su parte aérea espectacular, por lo que es fácil olvidarse del sistema radical que lo sostiene. Existen muchos comentarios sobre el limitado conocimiento acerca de las raíces de banano y plátano y de su ambiente. Sin embargo, quienes trabajan con los sistemas radicales, especialmente en la plantación, conocen el enorme esfuerzo que generalmente es necesario para extraer aún una pequeña cantidad de información.

Este volumen contiene los trabajos presentados en el Simposio internacional “Sistema Radical del Banano: hacia un mejor conocimiento para su manejo productivo”, realizado en San José, Costa Rica en noviembre de 2003. Aún cuando la salud de las raíces es una preocupación generalizada, la idea de realizar un simposio sobre este tema surgió como resultado de las inquietudes expresadas por el sector productor/exportador de banano, especialmente del señor Romano Orlich, Presidente de la Corporación Bananera Nacional de Costa Rica. CORBANA solicitó a INIBAP trabajar juntos para explorar soluciones a los problemas del sistema radical del banano. Este Simposio muestra que los sectores público y privado pueden compartir una agenda de investigación.

El objetivo del simposio fue presentar el conocimiento sobre el estado del arte del funcionamiento del sistema radical del género Musa y sus interrelaciones con microorganismos y plagas y proponer vías para mejorar la salud de la raíz. La demanda por este tipo de información aumenta cada vez más, debido a la necesidad de prevenir el deterioro en la salud del sistema radical que contribuye a una disminución constante en la producción de muchas fincas comerciales en América Latina y el resto del mundo. Las presentaciones cubren un amplio rango de niveles, desde células hasta plantas completas, y desde estudios de corto plazo hasta aquellos que reúnen información recolectada por muchos años y en muchos lugares. La apreciación de estas diferencias es importante ya que, por ejemplo, el conocimiento obtenido sobre los tejidos de las raíces en experimentos de corto plazo, debe ser integrado con el conocimiento a niveles más altos de organización en la planta y sobre escalas mayores de tiempo. De igual manera, el conocimiento obtenido en el laboratorio sobre la estructura y función de la raíz, o sobre las raíces y los microbios, debe ser comparado con los resultados obtenidos a nivel de campo. La única forma de incrementar nuestro conocimiento sobre los sistemas radicales del banano y el plátano, tanto en el laboratorio como en el campo, es facilitar las interacciones entre personas que trabajan a diferentes niveles. Esto es fascinante en sí mismo y además suministra las bases para tomar decisiones de manejo correctas, necesarias para mantener nuestros mercados y nuestro ambiente.

El trabajo sobre los sistemas radicales de banano y plátano se realiza usualmente como parte del amplio marco internacional de la investigación en sistemas de raíces de plantas. Este es un enfoque valioso. Se han logrado progresos significativos en cuando a evaluar el impacto de las plagas que atacan el sistema radical. Estamos aprendiendo sobre la importancia de tener un suelo sensible al manejo, tal como el drenaje, la irrigación y la fertilización. No debemos dejar este conocimiento de lado a medida que exploramos otros aspectos del desempeño de la raíz. Por otra parte, el progreso obtenido en el conocimiento del papel que juega la genética en el funcionamiento y la arquitectura del sistema radical del banano es poco. Efectivamente, aún debemos preguntarnos: ¿que clase de sistema radical queremos para las plantas de banano? La respuesta puede ser un tanto diferente, dependiendo de la situación local, pero hay algunos principios generales ya que las plantas siempre necesitarán agua, nutrientes y la habilidad de sostenerse en pie!. Estas respuestas pueden ayudarnos a ver claramente lo que es importante y a enfocarnos en cómo alcanzar nuestros objetivos. Durante la sesión final del simposio, los participantes elaboraron un perfil de lo que resta por hacer. Estos puntos representan el consenso de la diversidad de ideas e información presentados en el simposio.

Deseamos agradecer a todos los autores contribuyentes por su conocimiento del problema, a INIBAP y a CORBANA por reunirnos a todos y de esa forma hacer disponible esta publicación a escala mundial. Esperamos que estas Memorias estimularán a los lectores a explorar de que manera éste conocimiento podría traducirse en prácticas mejoradas de manejo.

David W. TurnerUniversity of Western Australia

Perth, AustraliaFranklin E. Rosales

Oficina Regional de INIBAP para América Latina y el CaribeTurrialba, Costa Rica


ObjectivePresent and discuss actual knowledge about and new insights into the functio-ning of the Musa root system, the inter-relation with micro-organisms and pests, and propose alternatives to improve root health.

ObjetivoPresentar y discutir el conocimiento actual y nuevos puntos de vista sobre el funcio-namiento del sistema radical de Musa, la interrelación con microorganismos y pla-gas, y presentar alternativas para mejorar la sanidad del sistema radical.

Organizing Committee / Comité organizador INIBAP-LAC Franklin E. Rosales Luis E. Pocasangre Thomas Moens Lissette VegaCATIE-INIBAP-LAC Galileo RivasCORBANA Jorge A. Sandoval Edgardo Serrano Mauricio Guzman Mario Araya Miguel GonzálezAssessor Ramiro Jaramillo

Administrative support / Apoyo administrativoINIBAP-LAC Lissette Vega CORBANA Sonia Jara

Compilation, translation and editing / Compilación, traducción y ediciónFranklin E. Rosales INIBAP-LACThomas Moens INIBAP-LACLissette Vega INIBAP-LACClaudine Picq INIBAP-HQ


1

Banana root deterioration and impacts on production

Deterioro radical del banano y sus impactos en la produccion


1 C. A Gauggel ([email protected]), G. Arévalo. College of Agricultural Science and Production. Zamorano Panamerican School-Honduras.

2 F. Sierra. Production Manager PROBAN .Colombia

The problem of banana root deterioration and its impact on production: Latin America’s experienceCarlos A. Gauggel1, Francisco Sierra2 and Gloria Arévalo1

AbstractThe deterioration of the banana root system and its effect on production is due to climatic, edaphic (physical and chemical) and biological factors. There are two kinds of root deterioration processes: 1) a fast one with accelerated collapse of the root system (days to months) due to site specific characteristics such as limited effective depth of soil, very high sand, gravel or clay content, high water table, and high soluble salt and sodium content; and, 2) a slow and gradual collapse which is due to the degradation of soil physical and chemical conditions, deterioration of soil biological activity, inefficient drainage (design, construction, management and maintenance) and poor crop management practices (disease control, nematodes, insects and weeds). The interaction between rainfall and topography ranks among the main climatic factors that can cause a fast or gradual collapse of the banana root system. On the other hand, corm rot (due to several causes), nematodes, and decrease in the soil biological activity are identified among the most important biological factors. The various physical, chemical and biological factors that cause deterioration of the root systems in Latin America are examined from the production point of view and experience of the authors. The application of new concepts and procedures (Soil Quality and Health) will make it possible to quantify the impact of the soil physical, chemical and biological properties in root performance and in banana production. This will also make it possible to determine, with precision, the production potential of the soils of the various banana zones and to manage them within the concepts and goals of sustainable production. The need to undertake more basic and applied research on the various factors that determine banana root performance are highlighted in order to design and implement more economically efficient production practices.

Resumen - La problemática del deterioro radical del banano y su impacto en la pro-ducción: Experiencia en América LatinaEl deterioro del sistema radical del banano y su efecto adverso en la producción se debe a factores climáticos, edáficos (físicos y químicos) y biológicos. Se presentan dos tipos de deterioro del sistema radical: 1) uno rápido con colapso acelerado del sistema radical (días a meses) debido a condiciones especificas al sitio tales como limitada profundidad efectiva del suelo, excesos de arena, grava o arcilla, nivel freático alto, y elevados contenidos de sales solubles y sodio; 2) un deterioro lento y gradual debido a la degradación paulatina de las propiedades físicas y químicas del suelo, deterioro de la actividad biológica, sistemas de drenajes deficientes (diseño, implementación, manejo y mantenimiento) y prácticas pobres de manejo del cultivo (control de enfermedades, nemátodos, insectos y malezas). Entre los factores climáticos se debe destacar la interacción de la precipitación y la topografía, la cual resulta en condiciones que pueden generar el colapso rápido o paulatino de la raíz. Por otro lado, entre los biológicos se destacan la podredumbre de los cormos (por diferentes causas), nemátodos y disminución de la actividad biológica del suelo. Las propiedades físicas, químicas y biológicas que comúnmente resultan en deterioro del sistema radical del banano en las diferentes áreas productoras de banano de América Latina son examinadas a la luz de la experiencia práctica de los autores realzando su impacto en la producción. La aplicación de nuevos conceptos y procedimientos (Calidad y Salud de Suelos) permitirán cuantificar el efecto de las propiedades físicas, químicas y biológicas del suelo en el desempeño de las raíces y en la producción. Esto también permitirá determinar con precisión el potencial productivo de los suelos de las diferentes


zonas bananeras y manejarlos dentro del concepto de producción sostenible. Se realza la necesidad de hacer más investigación básica y práctica en los diferentes factores que determinan el desempeño radical para poder diseñar e implementar prácticas de producción más económicamente eficientes.

IntroductionIt can be stated without a doubt that climate and soils determine the success of banana production enterprises. In most cases, climatic factors are easier to determine; howe-ver, the soil component is much more difficult to characterize due to the variation of the soil morphological, physical and chemical properties within a given area (large or small) under the same climate. The effect on banana root performance of the various components of these two factors has not been fully understood mainly due to the inte-ractions that occur among them. However, the effect of some soil properties on root performance has been understood to a considerable extent after the experience gained during the great expansion of the banana industry in the 1990s.

From an historical point of view, the understanding of the banana soil-plant rela-tionship can be divided into periods before and after 1990. Before 1990, the banana industry in general was very respectful of soil quality; only soils with optimum mor-phological, physical and chemical conditions were placed under banana production. However, after 1990 the lack of suitable soils for banana production, very high prices for the few available soils, decisions made without adequate soil studies, and dange-rous speculation typical of new and unknown markets, led many investors (large and small) to develop areas with undesirable soil conditions. This resulted in deterioration of the banana root systems in those areas, some gradual and some fast. In most cases these “experiments” associated with unfavorable market conditions had disastrous con-sequences for many investors. However, it also generated the need to better understand the causes and interrelations of the climate-soil-root complex in banana production. The level of knowledge of such a complex problem has increased significantly in the last thirteen years as a consequence of past experiences and the advent of new and pro-mising technologies. Thus, the objective of this paper is to undertake a general review of the production experience in Latin America as a consequence of banana root system deterioration in light of the lessons learned from the past and the opportunities offered by the future.

Causes of banana root system deteriorationIn general, the causes of banana root system deterioration are the same as those deter-mining yield. They can be classified as climatic, soil-related, and biological (diseases and insects affecting banana roots). The degree to which root deterioration occurs will be determined by the intensity of the predominant factors involved under the specific site conditions. In this way, the deterioration of the banana root system can be slow or fast - frequently too slow to be detected immediately.

Factors that cause a fast deterioration of the banana root systemThese factors are usually directly related to climatic conditions or soil properties and have drastic negative impacts on the banana plant. They are of major geographic importance in Latin America and strongly influence the banana root system (Table 1).


Table 1. Important factors involved in banana root deterioration and poor developmentFactors Effects in the banana root system

Soil factors

Restricted soil effective depth Weak development of the root system, shallow and thin roots

Extreme textures (massive clay, sand and burden) Weak root system, many dead roots, few live roots

Poor internal soil drainage. Short, weak and rotten roots, abundant dead roots

Soil compaction inherited from previous land uses Short, shallow and horizontal roots

High exchangeable and soluble sodium contents Few short functional roots with frequent injuries; many dead roots

High soluble salt concentrations Similar to above

Climatic and topographic factors

High water table Many dead roots, shallow root system

Frequent water logging of the upper soil horizons Many dead roots, many accidental roots, (rain and poor surface drainage) superficiaroot system

Frequent flooding Many dead and rotted roots, many adventitious roots

Phytopathological factors

Rotten roots Many dead roots, also corms with lesions

Climate and topographyRainfall is the most important factor involved in banana root system deterioration. It interacts with topographic factors that may result in severe adverse conditions for banana root development. The most important of the possible interactions are flooding, puddles after rains, shallow water tables (permanent or frequently fluctuating), and areas too close to sea level to be effectively drained.

In areas with rainy and humid climates, whose soils have high organic matter contents with high water retention, bananas generally experience root and corm rot. This results in a collapse of the plantation during the first crop and in an irreversible deteriora-tion of the root system. Such cases are common on the Atlantic coasts of Costa Rica, Panama and Colombia (specifically the Urabá zone) in areas previously used either for pasture, secondary forest or left fallow for long periods.

SoilsThe fast deterioration of the banana root system takes place when the soil has one or more of the following characteristics:

1. More than 60% coarse fragments by volume, high sand content (loamy sand or sand of coarse and very coarse size), very high clay content without soil structure (massive) or with coarse and very coarse blocks and prisms.

2. Effective soil depth less than 30 cm, restricted by continuous rock, massive clay or a shallow permanent water table.

3. High soluble salt concentrations with electrical conductivity greater than 4 dS/m.


4. High exchangeable sodium (PSI > 30%) and soluble sodium (RAS > 30) contents.

It is important to point out that these conditions can be devastating for the plant depen-ding upon the characteristics of rainfall and cultural practices. Severe dehydration of the plant occurs in areas with an intense and long dry season. This can be worse in plantations with poor irrigation systems on light-textured soils. On the other hand, soils with heavy, massive clays at shallow depths cause the quick collapse of the root system in climates with intense and long rainfall episodes. The damage is more severe if the drainage systems (surface and subsurface) are deficient. High soluble salt content is devastating for banana root systems in areas where evapotranspiration exceeds rainfall most of the year and water quality and availability are poor, such as in the banana zones of Santa Marta, Colombia, the Dominican Republic and the Southern Maracaibo Lake of Venezuela. In several cases, exceptionally good climatic conditions or costly efforts to supply excess irrigation water for leaching of salts, may temporarily delay the collapse. However, when the weather returns to normal or efforts to avoid the problem diminish, the collapse of the roots occurs rapidly. Great efforts to improve this situation are only possible during times of favorable markets. The first half of the 1990s is a good example of this.

Biological factorsAreas with high nematode populations, other banana root parasitic microorganisms and insects can cause fast banana root deterioration, especially when the areas have been previously planted with bananas.

Factors that cause the gradual deterioration of the banana root systemThese factors can also be classified as climatic, soil-related (edaphic) and biological with very complex interrelations depending upon specific site conditions. Their impact on yields depends on the intensity at which the limiting factor occurs at the given site (Table 2).

Climatic factorsExtreme climatic conditions (severe drought or heavy rainfall), without effective measures to protect the banana root system, can result in its gradual deterioration. In areas subject to severe drought, the deterioration of the root system is a consequence of either a lack of, or an inadequate, irrigation system and an insufficient supply of good quality water. This condition is very frequent in the Santa Marta banana zone. If these conditions are light to moderate the root system deterioration is gradual; however, under severe conditions the collapse is fast.

In areas with high and fairly even rainfall patterns (Atlantic coasts of Costa Rica, Panama, and Urabá, Colombia) the deterioration of the banana plant root system is gra-dual. Such conditions occur where there is an inefficient drainage system, the depth to the water table is less than 0.9 m (this condition may occur also in dry areas) or perched water tables within the upper 0.9 m as a consequence of impermeable soil horizons. This latter condition is very common in high rainfall areas. It can also occur in dry areas as a consequence of applying irrigation water at higher volumes than required.


Table 2. Important factors involved in the gradual deterioration on the banana root system.Factor Effect in the root Potential reduction of maximum performance (%)

Soil factors

Organic matter loss Poorly developed root system Variable

Soil erosion Poorly developed root system Variable

Soil salinization by saline Poorly developed root system 30-80irrigation water with injuries

Acidification by excessive Poorly developed root system 10-50fertilizer use with injuries

Loss due to porosity and Few, short and thin roots 10-60soil compaction

Low fertilizer rates Poorly developed root system Variable

High permanent and perched Short, rotten roots and high 10-60water tables root mortality

Poor surface drainage Rotten and high mortality and 10-70 adventitious roots

Nutrient imbalances Poorly developed root system Variable

Inversion of soil profile Poor developed root system, 0-50 adventitious roots

Biological factors

Increase in nematode population 70

Decrease in soil micro and Poorly developed root system Variablemacroorganism populations

During long and intense rainfall (cold fronts in the Atlantic coasts of Central America), despite adequate drainage systems, the amount and duration of rainfall saturates the soil causing anaerobic conditions in the root zone. Such conditions result in the forma-tion of abundant adventitious roots, deterioration of the underground root system, and temporary reductions in yield and fruit quality.

Soil factors

Effective soil depth and soil textureLimitations in effective soil depth between 30 and 90 cm usually result in gradual deterioration of the root system depending upon what constitutes the limiting physical barrier (coarse fragments, sand or massive clay) and the texture of the horizon itself. If massive heavy clay is limiting, rainfall or excessive irrigation can cause temporary water logging in the root zone resulting in damage to the root system. On the other hand, predominantly sandy textures result in frequent water deficits or nutrient lea-ching that in turn, weaken the root system.

Soil erosion, organic matter loss and soil compactionTo varying degrees, the soils of the banana farms are subject to water erosion. The main causes are high intensity rainfall and sprinkler irrigation (under and above tree). Soil


cover is of paramount importance to prevent significant soil losses. Weeds under pro-per control can be a great ally in this task. Soil erosion causes substantial losses of soil organic matter, since soil colloids (humus and clays) are the most vulnerable to loss. Compaction of the topsoil horizon is an almost unavoidable process due to frequent human traffic for various agricultural practices. Such compaction never reaches deeper than 15 cm, although, the loss of porosity at the soil surface reduces oxygen supply and soil infiltration rate. These processes take place slowly over months, and even years, and the effect is not noticeable in the short-term. Likewise, root deterioration takes place gradually over time resulting in weak root systems that eventually affect yields. Depending on the time elapsed, and the intensity of the causal factor, the reduction in yield can be severe.

Soil management practices (fertilization, irrigation and drainage)The production experience indicates that poor agricultural practices can cause dete-rioration of the banana root system. Inadequate (too low or too high) and inefficient irrigation and drainage should be highlighted. These factors cause a significant deterio-ration of the banana root system and in general are a consequence of restricted budgets, the lack of proper diagnostic techniques and poor maintenance of the irrigation and drainage systems. Inadequate fertilization practices cause nutritional deficiencies and imbalances that, in turn, also cause root deterioration.

Inversion of the soil profileThe frequent maintenance of the drainage system requires spreading the sediment dug out from the bottom of drainage canals on the soil surface. Most of this sediment comes from the walls of the drainage canals and the lower soil horizons. Over time, the sedi-ment spread on the surface reaches significant thickness, greater than 50 cm. In cases where the sediment is too clayey or sandy, soil quality diminishes substantially. In clayey soils the infiltration rate decreases substantially as a result of the loss of porosity in this new soil horizon with intense human traffic. This condition presents an obstacle to the flux of water and gases between the atmosphere and the soil and also for water infiltration, thus aggravating surface drainage conditions (Table 2).

Biological factorsNematodesIncreases in nematode population and other banana root pathogens are, without a doubt, one of the most critical factors affecting the gradual deterioration of the banana root system. Such conditions can become chronic and very serious if not managed. To maintain a vigorous root system in banana plantations requires careful pest and disease control, taking into consideration innovative alternatives that promote biological diver-sity and healthy activity of beneficial microorganisms in the soil environment.

Beneficial soil microorganismsThe excessive, but sometimes necessary, use of pesticides, poor irrigation water quality, excessive use of fertilizers, soil erosion, loss of soil organic matter and soil compaction, all upset the soil population of beneficial microorganisms. This is a field that is recognized, at least conceptually, by the banana industry to be very important but has lacked scientific research in banana plantations. It is very important for the


banana industry to undertake vigorous research in this area to determine the effect of the decrease and changes in the populations of soil organisms, especially micro-orga-nisms, on yields.

Interrelation between production and the deterioration of the banana root systemSoil physical propertiesThe production experience through time indicates that those physical properties that limit the development and performance of the banana root system, have considerable impact on yields and in many instances lead to production lower than that required for profitable returns (Table 3). As previously indicated, the intensity of the specific limi-ting factor will determine the crop yield. Some of these limiting factors cannot be signi-ficantly improved, for example soil texture. In other cases, however, the improvement will depend on the cause of the limiting factor. For example, effective soil depth can be improved by soil rehabilitation practices. The cost of rehabilitation will depend on the characteristics of the soil-limiting factor (for example, rock, gravel, sand, clay).

Soil chemical propertiesOf the range of chemical properties that restrict the development and performance of banana roots, the best understood are the percentage of exchangeable sodium, the sodium absorption ratio and the soluble salt content. These conditions generally occur in areas where evapotranspiration exceeds rainfall significantly, as in the banana zones of Santa Marta, Colombia, some areas of the Dominican Republic or where elevation above sea level and the effect of tides do not allow efficient leaching of soluble salts and sodium. In the cases where the concentrations of soluble salts and sodium are high enough, the collapse of the banana root system is very fast; these areas are usually abandoned after the first few crops. Some small areas that are affected by high soluble and exchangeable sodium occur within large non-affected areas, and they are kept under production. Such areas have low yields and are kept under production at a very high cost and constitute a burden for the farm that contains them.

Table 3. Common banana crop yields in areas with physical properties that affect the performance of the banana root system. Mean yields for different areas of Latin America.Physical factors restrictive for banana root Mean annual productiondevelopment boxes ha-1 year-1

Effective soil depth 1200-2000

Texture 1100-2200

Shallow soil compaction 1200-1800

Low available water 1000-1900

Poor soil internal drainage 900-1700

Flooding 1200-2000


Areas with soils with high iron and manganese concentrations and exchangeable alu-minum usually have low yields (Table 4). Field observations indicate short and thin roots with dark brown and black lesions on the epidermis. Whether this is a direct effect of elemental toxicity on the roots or the effect of interactions with multiple factors, has not been determined since the soils with these characteristics are very low in phosphorus, cations, zinc and boron. In most cases, areas with these characteristics are kept under production at high cost. Lime and dolomite applications have given partially effective results, probably due to application rates that are too low for the soil requirements.

Table 4. Common banana plantation yields in areas severely affected by soil chemical factorsthat affec root performance. Average yield ranks for different areas of Latin America.Restrictive factors for root development Annual average yield boxes ha-1 year-1

Exchangeable sodium percentage 900-1500

Soluble salts 700-1200

Soluble iron and manganese 1200-1700

Exchangeable aluminium 1200-1700

Soil quality and healthThese are two very important concepts that have been developed in the last decade in Soil Science and without doubt will help us to better understand and quantify the interrelation of the soil properties and the banana root system and production under the perspective of sustainable agriculture.

Soil qualitySoil quality is defined as the capability of the soil to function effectively in the present and future. This integrates physical, chemical and biological soil processes establishing the most relevant for the production of biomass of sustainable quality necessary to generate good plant and animal health (Doran and Parkin 1994). The quantification of the effect of soil in biomass production will depend on the impact of each individual soil property on the performance of the crop of interest. The concept applied by Karlen and Stott (1994), to assign weightings to the relevant soil properties involved in the effects of soil erosion, was applied to evaluate the effect of these properties in the production of crops other than bananas (Barahona 2000). The success of the practical application of these concepts (Fernandez 2003, Cueva 2003, Orellana 2003) and their usefulness to predict the performance of several crops has lead to the application of these concepts to banana cultivation. Considering the soil requirements for the adequate performance of the banana root system and its impact on crop production, weights can be assigned to the key soil characteristics according to their role in banana production (Table 5). Such an approach makes it possible to quantify the effect of these key soil properties and to correlate them with yields in a quantitative manner. According to their importance in


determining banana plant yield, soil effective depth, soil texture, soluble salt concen-trations, and the various forms of sodium in the soil, have the highest weights.

Soil healthBesides the soil morphological, physical and chemical characteristics, soil health emphasizes the biological activities in the soil (Roming et al. 1996). In Latin America, the application of this concept makes it possible to create a system to develop the potential of the soil that is more adapted to the technical expertise of small farmers, emphasizing the sustainability of the soil resource. These are methods adapted for use by farmers with little formal education and technical knowledge. These methods will allow simple but effective field evaluations of the interrelations between soil and plants in a continuous way at a low cost (Roming et al. 1996, the Cornell University, Zamorano and Soil Working Group 2002). With some modifications, these methods will be of much help for banana and plantain producers focusing efforts on the conti-nuous improvement of the root environment.

ConclusionsBanana plantation yield is a direct function of the physical, chemical and biological conditions under which the banana plant roots develop and the interrelations that occur between climate and cultural practices. The deterioration of the banana root system can be quick or slow, often imperceptible in the short term. The intensity of the various factors involved in the deterioration of the banana root system varies throughout the different banana producing zones of Latin America. These conditions are usually the same in the different geographic locations except where climate and elevation above sea level are the main causes.

Among the most important soil factors that result in a fast deterioration of the banana root system are severely restricted effective soil depth, extremes in soil texture (sand or clay) and the soluble salt and sodium and exchangeable sodium concentrations. Lower concentrations of soluble salts and sodium, low nutrient concentrations, poor irrigation and drainage practices, low populations of soil organisms that are beneficial to the soil, high numbers of nematodes, diseases of the root system and poor agricultural practices

Table 5. Soil quality indices of morphological, physical and chemical soil properties that determine banana production.Physical /morphological Weight (0-1) for Chemical property Weight (0-1) for property plant development plant development

Effective soil depth 0.70 pH 0.25

Texture 0.60 Base saturation 0.20

Structure 0.35 Organic matter 0.40

Internal soil drainage 0.35 Exchangeable aluminum 0.40

Resistance to penetration 0.35 Soluble salts 0.60

Hydraulic conductivity 0.35 Sodium absorption ratio (SAR) 0.60

Bulk density 0.35 Exchangeable sodium percentage 0.60

Available water 0.50 Cation exchange capacity 0.40

The problem of banana root deterioration and its impact on production22 C.A. Gauggel et al. 23 D.H. Marin (abstract) 23

(inversion of the soil profile) generally result in slow and progressive deterioration of the banana root system.

The production experience indicates that yields are limited by conditions that weaken the banana root system. Ignoring the optimum conditions for banana plant roots has lead to substantial economic losses, either in an abrupt or progressive manner, depen-ding upon the degree of impact. Given the economic significance of the issues dis-cussed here, it is of paramount importance to conduct more basic and applied research on the interactions among the factors that limit banana root system performance. The practical application of new concepts and technologies will be of great importance to make effective business decisions.

ReferencesBarahona R. 2000. Caracterización detallada de los suelos de San Nicolás y practicas recomendadas para su

sostenibilidad. El Zamorano, Honduras. Undergraduate Thesis. Pan-American School of Agriculture. El Zamorano. Honduras.

Cueva F. 2003. Caracterización de suelos y fertilización de frutales de altitud media alta en el Occidente de Honduras. Undergraduate Thesis. Pan-American School of Agriculture. El Zamorano. Honduras.

Doran J.W. & D.C. Parkin. 1994. Defining and assessing soil quality. Pp. 6, 9 and 11 in Defining soil quality for a sustainable environment. (J.W. Doran, J.A.E. Molina & R.F. Harris, eds). Soil Science Society of America (SSSA). Special Publication No. 35. Madison, WI.

Fernández J. 2003. Caracterización detallada de suelos de los sectores de Zorrales, Monte Redondo de El Zamorano para el establecimiento y renovación de pasturas. Undergraduate Thesis. Pan-American School of Agriculture. El zamorano, Honduras.

Karlen D.E. & D.E. Stott. 1994. A framework for evaluating physical and chemical indicators of soil quality. Pp. 61-63 in Defining soil quality for a sustainable environment. (J.W. Doran, J.A.E. Molina & R.F. Harris, eds). Soil Science Society of America (SSSA). Special Publication No. 35. Madison, WI.

Orellana S. 2003. Caracterización de suelos y fertilización para frutales de clima templado en el Occidente de Honduras. Undergraduate Thesis. Pan-American School of Agriculture, El Zamorano, Honduras.

Roming D.E., M.J. Garlyn & R.F. Harris. 1996. Farmer-based assessment of soil quality a soil health scorecard. Pp. 39 and 49-59 in Defining soil quality for a sustainable environment. (J.W. Doran, J.A.E. Molina & R.F. Harris, eds). Soil Science Society of America (SSSA). Special Publication No. 35. Madison, WI.

Cornell University, Zamorano, Soils Working Group. 2002. Guía de Salud de Suelos. Pp. 45-55 in Manual para el cuidado de la salud de suelos. Zamorano Academic Press, Honduras.

The problem of banana root deterioration and its impact on production22 C.A. Gauggel et al. 23 D.H. Marin (abstract) 23

Research in progress and future perspectives on the root system managementDouglas H. Marín1

AbstractPlant-parasitic nematodes are the most important pests affecting the root system of bananas. Although poly-specific communities are normally found, the burrowing nematode, Radopholus similis, is usually the dominant species. Nematode management has mainly relied on the application of chemical granular nematicides, in addition to some cultural practices such as the propping and-or guying of fruit-bearing plants. Carbamate and organophosphate nematicides are usually rotated two to three times per year under Costa Rican conditions. However, root growth has not been improved in recent years in spite of nematicide treatments. Because root decay is not only associated with nematode damage, current research has also focused on other micro-organisms and soil conditions to evaluate their effect on root growth and its deterioration. Crop rotation, fallow, and organic amendments have been evaluated without success. Products from biological and/or organic origin were found to be available but no commercial application has been implemented yet. Environmental concerns and worker safety and health make the future of chemical nematicides uncertain; therefore, new ideas for nematode management are required to control these microorganisms in areas such as Costa Rica, where burrowing nematode is one of the main production constraints. Further research on biological control and non-chemical alternatives is required. Soil amendments to improve competition, predation and/or antibiosis may also play an important role on root growth. Genetic variation of the nematode population is also another aspect that may bring some light to understand the complex puzzle of improving banana root growth under tropical conditions.

Resumen - Investigaciones en progreso y perspectivas futuras para el manejo del sistema radicalLos nematodos fitoparásitos son las plagas más importantes que afectan el sistema radicular del banano. Aunque las poblaciones son normalmente poli-específicas, el nematodo barrenador, Radopholus similis, es usualmente la especie dominante. El manejo de los nematodos ha dependido principalmente de la aplicación de nematicidas químicos granulares; además de algunas prácticas culturales complementarias como el apuntalamiento de las plantas fructificadas. Nematicidas carbamatos y organofosforados son usualmente rotados dos o tres veces al año bajo las condiciones de Costa Rica. Sin embargo, el crecimiento radicular no ha mejorado en los últimos años, a pesar de los tratamientos nematicidas. Debido a que el deterioro radicular no sólo está asociado al daño causado por los nematodos, la investigación actual ha enfocado su atención a otros microorganismos y/o condiciones de suelo para evaluar su impacto en el crecimiento de las raíces y su destrucción. La rotación de cultivos, el barbecho y la incorporación de enmiendas orgánicas se ha evaluado sin éxito sostenible. Productos de origen biológico u orgánico también se han encontrado disponibles pero no se ha documentado su uso a nivel comercial. Las preocupaciones desde el punto de vista ambiental y de salud de los trabajadores hacen que el futuro de los nematicidas químicos sea incierto. Nuevas ideas en el manejo de nematodos se requieren para el control de estos organismos, en áreas como en Costa Rica, donde el daño causado por el nematodo barrenador es una importante limitación para la producción de banano de exportación. Mayor investigación en control biológico y alternativas no químicas también se necesitan. Enmiendas al suelo para mejorar la competencia, la depredación y la antibiosis también pueden tener un rol importante en el crecimiento radicular. La variación genética de las poblaciones de nematodos es también otro aspecto a considerar para el entendimiento del complicado rompecabezas que es el mejoramiento del sistema radicular de banano bajo condiciones tropicales.

1 Del Monte, P.O. Box 484-1000, San José, Costa Rica

M. Castro et al. (abstract)24 C.A. Gauggel et al. 25E. Serrano 25

Banana yield response to nematode control in Dole farms in Costa RicaMarco Castro1, Ricardo Duarte2, Carlos Portillo3, Jorge Gonzales4

AbstractThere are various biotic and abiotic factors that may affect banana root development. They include soil temperature, soil texture, soil compaction, drainage, nutrition, mineral toxicity, fungi, weevils and nematodes. The objective of this presentation is to document the yield response of banana to nematode control with conventional chemical control programs. In Central America, banana yield reduction has been related to infestation of Radopholus similis, the effect of other nematode genera on yield has not been documented. Nematode management has been evaluated experimentally with various cultural, microbial, botanical and chemical control alternatives. Current commercial management of R. similis and its damage includes cultural practices such as propping, drainage, organic amendments, fallow, rotation and tillage. Additionally, the use of conventional chemical nematicides in the organophosphate and carbamate groups is required under high nematode pressure environments such as in Costa Rica. Results presented are based on 6 experiments conducted in the Rio Frio and Estrella Valley zones in Costa Rica. Yield (stems harvested, stem weight, calibration, finger length, tons per hectare and boxes per hectare) and root health data (R. similis numbers, percentage of functional roots and fresh weight of functional roots) has been gathered from replicated plots over several years. The information to be presented compares yields and root health with a commercial chemical control programme versus an untreated control. Reduction for all yield components was measured as a result of nematode infestation, in general differences started at 9-12 months after stopping chemical control treatments. The yield loss was measured in all experimental sites with reduction in total yield ranging from 30-50% in the absence of nematode control.

Resumen - Respuesta en producción de banano al control de nematodos en fincas de Dole en Costa RicaExisten varios factores bióticos y abióticos que podrían afectar el desarrollo de las raíces del banano incluyendo: temperatura del suelo, textura del suelo, compactación del suelo, drenaje, nutrición, toxicidad mineral, hongos, picudos y nematodos. El objetivo de esta presentación es documentar la respuesta de la producción de banano al control de nematodos con programas de control químico convencionales. En América Central, se ha relacionado la reducción de la producción de banano con la infestación de Radopholus similis; el efecto de otros géneros de nematodos no ha sido documentado. El manejo de nematodos ha sido evaluado experimentalmente con varias alternativas culturales, microbiológicas, botánicas y de control químico. El manejo comercial actual de R. similis y su daño incluye prácticas culturales como el apuntalamiento, drenaje, enmiendas orgánicas, barbecho, rotación y labranza. Adicionalmente, el uso de nematicidas químicos convencionales es necesario en ambientes con alta presión de nematodos como en Costa Rica. Los resultados presentados se basan en 6 experimentos efectuados en las zonas de Río Frío y del Valle de la Estrella en Costa Rica. El rendimiento (tallos cosechados, peso del tallo, longitud del dedo, toneladas por hectárea y cajas por hectárea) y datos sobre la salud de la raíz (número de R. similis, porcentaje de raíces funcionales y gramos de raíces funcionales) se han recolectado de parcelas repetidas en diferentes años. La información a presentar compara el rendimiento y la sanidad radical entre un programa de control químico comercial y uno sin tratamiento. Se midió la reducción de todos los componentes de rendimiento como resultado de la infestación con nematodos. En general las diferencias comenzaron a partir de 9 a 12 meses después de terminar los tratamientos químicos de control. La pérdida de rendimiento se midió en todos los sitios experimentales con una reducción en el rendimiento total entre 30 y 50% en ausencia de control de nematodos.

1 Research Director Dole Fresh Fruit de Honduras S.A., Colonia El Naranjal, La Ceiba, Honduras, e-mail: [email protected]; 2 Research Manager Standard Fruit Costa Rica, P.O. Box 12-1007, San José, Costa Rica, e-mail: [email protected]; 3 Research Assistant Standard Fruit Honduras, e-mail: [email protected]; 4 Vice President Agricultural Research Dole Food Company, San José, Costa Rica, e-mail: [email protected]

M. Castro et al. (abstract)24 C.A. Gauggel et al. 25E. Serrano 25

Relationship between functional root content and banana yield in Costa RicaEdgardo Serrano1

AbstractThe study examines the relationship between functional root content and yield over the 16 years from 1987 to 2002. The counties studied were Sarapiquí, Pococí, Guácimo (production areas west of the Reventazón river), and Siquirres, Matina and Talamanca (production areas east of the Reventazón river). Banana plantations cover 38 558 ha in these counties and represent 91% of the banana crop planted in Costa Rica. Functional root fresh weights (g/plant) ranged from 38 to 87 in Sarapiquí; 43 to 102 in Pococí; 36 to 81 in Guácimo; 36 to 135 in Siquirres; 35 to 143 in Matina and 31 to 114 in Talamanca counties. The historical production, in boxes of 18.14 kg ha-1 year-1, varied annually from 2019 to 2864 in Sarapiquí; 1778 to 2643 in Pococí; 1723 to 2536 in Guácimo; 2271 to 2947 in Siquirres; 2228 to 2581 in Matina and 1740 to 2984 in Talamanca counties, according to CORBANA S.A. The best yields per county were achieved in 1989: 2643 boxes ha-1 year-1 in Pococí, 2947 in Siquirres and 2740 in Talamanca counties. These values coincide with functional root fresh weights (g/plant) of 102 for Pococí, 135 for Siquirres and 100 for Talamanca. A significant relationship between functional root weights and productivity was found for Sarapiquí (y = 5.98x + 1922; P<0.036); Pococí (y = 14.05x + 1104; P<0.0001); Siquirres (y = 5.48x + 2065; P<0.0001) and Talamanca (y = 9.96x + 1922; P<0.0002) counties. Results showed that for every 10 g of functional roots lost, productivity (boxes ha-1 year-1) decreased by 60 in Sarapiquí, 140 in Pococí, 55 in Siquirres and 99 in Talamanca counties. Biotic and abiotic factors that can affect functional root content and productivity in Costa Rica are highlighted.

Resumen - Relación entre el contenido de raíz funcional y la producción de banano en Costa RicaSe estudió la relación entre el contenido de raíz funcional y la productividad de 16 años (de 1987 a 2002). Dicho estudio se realizó en los cantones de Sarapiquí, Pococí y Guácimo de la zona de producción al oeste del río Reventazón y Siquirres, Matina y Talamanca de la zona de producción al este del río Reventazón. En conjunto el área suma 38.558 ha equivalente al 91% del área total sembrada en Costa Rica. Los pesos de raíz funcional (g/planta) oscilaron de 38 a 87; 43 a 102; 36 a 85; 36 a 135; 35 a 143 y 31 a 114 en los cantones de Sarapiquí, Pococí, Guácimo, Siquirres, Matina y Talamanca, respectivamente. Los datos promedio de productividad de dichos cantones, según las Estadísticas de Exportación Bananera de Costa Rica revelan rendimientos (cajas de 18,14 kg ha-1 año-1) muy variables en los diferentes años y cantones, que fluctuaron de 2019 a 2864 en Sarapiquí; 1778 a 2643 en Pococí; 1723 a 2536 en Guácimo; 2271 a 2947 en Siquirres; 2228 a 2581 en Matina y 1740 a 2984 en Talamanca. La productividad por cantón muestra que 1989 fue el año con los mejores rendimientos, 2643; 2947 y 2740 cajas ha-1 año-1 coincidiendo dichos valores con pesos adecuados de raíz funcional de 102, 135, y 100 g por planta para los cantones de Pococí, Siquirres y Talamanca. La relación entre raíz funcional y productividad fue significativa para los cantones de: Sarapiquí (y = 5.98x + 1922; P<0.036); Pococí (y = 14.05x + 1104; P<0.0001); Siquirres (y = 5.48x + 2065; P<0.0001) y Talamanca (y = 9.96x + 1922; P<0.0002). Las ecuaciones establecen que por cada 10 g de pérdida de raíz funcional se disminuye la productividad en 60, 140, 55 y 99 cajas ha-1 año-1 en los cantones de Sarapiquí, Pococí, Siquirres y Talamanca, respectivamente. Se señalan factores bióticos y abióticos que pueden haber afectado el contenido de raíz funcional y por ende la productividad.

1 Dirección de Investigaciones, CORBANA. Apdo.390-7210 Guápiles, Costa Rica. E-mail : [email protected]

The problem of banana root deterioration and its impact on production26 Relatioship between functional root content and banana yield in Costa Rica26 C.A. Gauggel et al. 27E. Serrano 27

Introduction Variations in Costa Rican banana yields (boxes ha-1 year-1) were studied from 1987 to 2002. A combination of factors has affected banana yield including functional root fresh weight, which, in turn, is affected by biotic factors such as nematodes and soil micro-organisms, and abiotic factors such as superficial and internal drainage, element toxicity (Al+3, Fe+3 and Mn+2), high precipitation, low and high temperatures, soil compression and cultivation, and light or heavy-textured soils.

Serrano and Marin (1998) correlated functional root weight with annual yield, calcu-lated in boxes ha-1 year-1 and boxes/processed bunch from 1987 to 1997, and found correlation coefficients of 0.85 and 0.92, respectively. When the total and functional root weights were determined in the different banana producing counties in Costa Rica from 1995 to 1999, it was found that farms with yields close to 3000 boxes/ha/yr had higher total and functional root weights than farms with lower yields (Calvo and Araya 2001). This study relates functional root fresh weight (taken by a standard sampling procedure) with yield/county from 1987 to 2002 and highlights historical biotic and abiotic factors that directly or indirectly affected functional root fresh weight.

Historical yield at national levelThe data are from farms located in the Sarapiquí, Pococí, Guácimo, Siquirres, Matina and Talamanca counties, which represent 38 558 ha, or 91% of the planted area. From 1987 to 2002 there were great variations in the yield measured in boxes ha-1 year-1 in Costa Rica. The total area planted with banana in 1987 was 20 987 ha with an average yield of 2476 boxes of 18.14 kg ha-1 year-1 This increased to 2731 boxes ha-1 year-1 in 1989, the highest yield obtained in the last 16 years (Figure 1). From 1989 to 1992, banana yields declined continuously, reaching a low of 2400 boxes ha-1 year-1 Since

Figure 1. Changes in productivity and in the surface area of bananas planted in Costa Rica, 1987-2002.

1500

2000

2500

3000

1987 1989 1991 1993 1995Year

1997 1999 2001

(Box

es/h

a)

0

10000

20000

30000

40000

50000

60000

(ha)

Annu

alprod

uction

YieldAccumulated area planted after 1987Total production area


then, some years showed limited recovery, such as 1998 where yield reached 2466 boxes/ha/yr but other years saw even lower yields, such as in 2002, which had the third lowest yield of the period with 2107 boxes ha-1 year-1 for the 42 182 ha planted.

Relationship between functional root fresh weight and yield per countyProduction area west of the Reventazón riverWest of the river are the counties of Sarapiquí, Pococí and Guácimo. Pococí county is the largest producer in Costa Rica. It has maintained a production area of 10,169 ha and has exported more 18.14 kg boxes of bananas than any other county over the last 16 years, reaching 340 million boxes (Sanchez and Zuñiga 2003). Functional root fresh weights, taken using a standard sampling technique (Araya, these proceedings) were between 43 and 102 g/plant, and the lowest and highest weights coincided with the lowest and highest yields (Figure 2). The data from 1995 to 2002 in Sarapiquí county showed trends very close to those in Pococí (Figure 2). Functional root weights were between 38 and 87 g/plant and the yields for these years were between 2084 and 2535 boxes. The regression between functional root weight and productivity was highly significant in Pococí county (P>0.0001, y = 14.06x + 1104 R2 = 0.77) as well as in Sarapiquí (P = 0.036, y = 5.98x + 1922 R2 = 0.55). Thus, productivity on the Pococí farms is much more sensitive to loss of functional roots than in Sarapiquí. For every 10 g reduction of root there is a loss of 140 boxes ha-1 year-1 in Pococí and and 60 in Sarapiquí. Although Guácimo county followed the same trend, the results were not significant, probably because fewer farms had historical functional root weight data.

Production area east of the Reventazón riverEast of the river are Siquirres, Matina and Talamanca counties. During the period, 248 million 18.14 kg boxes were exported from Siquirres county, which maintains an ave-rage production area of 7 033 ha. Siquirres is the best producer east of the Reventazón River and second best nationally. Functional root weights (g/plant) in Siquirres were 36 to 135; in Matina, 35 to 143; and in Talamanca, 31 to 114. The corresponding annual yields (boxes/ha) were 2271 to 2947 for Siquirres; 2228 to 2581 for Matina and 1749 to 2984 for Talamanca (Figure 3). Similar to the western counties, highest and lowest yields corresponded with highest and lowest functional root fresh weights. The rela-tionship between functional root weight and productivity was highly significant for Siquirres county (P>0.0001, y = 5.48x + 2065 R2 = 0.79) and Talamanca (P>0.0002, y = 9.96x + 1922 R2 = 0.64). For each 10 g of functional root weight that was lost, annual productivity decreased by 55 boxes ha-1 year-1 in Siquirres and 99 in Talamanca county. Although the trend was the same in Matina county, the results were not significant.

Biotic and abiotic factors affecting banana productivity in Costa Rica from 1987 to 2002To understand the relationship between national productivity and functional root fresh weight in Costa Rica in the study period, it is necessary to review banana indus-try activity. From 1987 to 1992, productivity in Costa Rica was acceptable, due to


1600

2000

2400

2800

Boxes/ha

30

50

70

90

110g

Functional roots (g) Yield

Sucker survey critical level Mother sucker survey critical level

Pococí

1800

2050

2300

2550

2800

1987 1989 1991 1993 1995 1997 1999 2001Year

25

45

65

85

105

125

Sarapiquí

1500

2000

2500

30

50

70

90

Guácimo

Figure 2. Functional root fresh weight (g/plant) and productivity (boxes ha-1 year-1) from 1987 to 2001 for three counties west of the Reventazón River. The mother sucker survey critical weight is 60 g and the sucker survey critical weight is 42 g (Calvo and Araya 2001).


Functional roots (g) Yield

Sucker survey critical level Mother sucker survey critical level

1900

2150

2400

2650

30

70

110

150

Matina

1700

2200

2700

1987 1989 1991 1993 1995 1997 1999 2001Year

25

45

65

85

105

125

Talamanca

2200

2450

2700

2950

Boxes/ha

20

60

100

140g

Siquirres

Figure 3. Functional root fresh weight (g/plant) and productivity (boxes ha-1 year-1) from 1987 to 2001 for three coun-ties east of the Reventazón River.


favourable biotic and abiotic factors. The fenamiphos nematicide was replaced in 1987 by new compounds such as terbuphos, aldicarb and ethoprop. The constant and indis-criminate use of fenamiphos from the beginning to the middle of the 1980s probably caused the accelerated biodegradation observed (Durán 1988). The introduction of the new compounds gave better control of nematodes, and consequently healthier root systems associated with better productivity. Propiconazole, introduced into the market in 1987, and other systemic fungicides like tridemorph and benomyl achieved better black Sigatoka control (Guzmán 2003). Finally, well-distributed precipitation with no excessive flooding or lengthy droughts, especially during 1987, 1988 and 1989, and markets with no “distortions” (no fruit export restrictions) contributed to this period of good productivity.

During 1993 and 1994, productivity dropped dramatically to 2046 and 1960 boxes ha-1 year-1, respectively. Farmers did not expect this decrease in yield. During the last Banana Development Programme between 1989 and 1994 about 30 000 addi-tional hectares of banana were planted, increasing the area from 22 022 ha in 1988 and to 52 737 ha in 1994. This expansion can be explained by the opening of new markets in the Eastern European countries. An increase in productivity was expected (Serrano and Marín 1998), but did not occur, probably because of the combination of various factors including aldicarb, which was effective, but was taken off the market in 1990, reducing the options for rotating terbuphos, ethoprop, carbofuran and oxamyl. With this, the options for controlling Radopholus similis decreased and the risk of inducing accelerated biodegradation of the nematicides increased (Anderson et al. 1998). This phenomenon could have been accelerated in banana soils west of the Reventazón River where, historically, one more dose of nematicide was applied per year compared with the eastern areas. In general, these soils have a lighter texture and a lower pH than the soils in the east. Under these favorable conditions, R. similis and Meloidogyne incognita (Davide 1980) caused more serious banana root damage. At the end of 1992, Mycosphaerella fijiensis showed decreased sensitivity to propiconazole, which resul-ted in fewer leaves at harvest and lower bunch weights (Romero and Sutton 1997, Serrano and Marin 1998, Guzmán 2003).

The globalization of the banana market is without doubt one cause of the low produc-tivity during this period. This was manifested in two ways, 1) High global demand for fruit caused by the expectation of new markets in Eastern Europe which did not mate-rialize and, 2) European Union protection of its ex-colonies, which negatively affected the export of Latin American bananas to Europe (Zuñiga 1994). The aforementioned factors decreased the liquidity of independent growers, which resulted in cost res-tructuring including a decrease in farm maintenance. This accelerated the decrease in productivity in 1994.

Low precipitation in 1995 began the slow recovery of productivity from 1995 to 1998 (Figure 4). Another factor that helped the recovery was the introduction of a cadusaphos nematicide. However, because of its high cost, it was not commonly used. Fenamiphos was reintroduced on the banana market (Navarro 1995) with manufacturer’s instruc-tions to apply only once a year to avoid accelerated biodegradation.

The “Framework Agreement” with the European Union (Zuñiga 1994) improved the banana market in Europe. New options for black Sigatoka control increased the number


Figure 4. Annual precipitation in six banana-growing counties in the Caribbean coast of Costa Rica, 1987-2002.

of leaves at harvest and bunch weight as well. Other triazole compounds were introdu-ced to the market in 1996 and azoxistrobin was introduced in 1997. Azoxistrobin is one of the new group of systemic fungicides called strobilurins and has effects comparable with those of propiconazole (Guzman and Romero 1997, Serrano and Marin 1998). In addition to the use of new fungicides, renewed relationships with transnational compa-nies and with independent growers, and optimal climatic conditions during the second semester of 1997 and most of 1998, made possible the productivity increase to 2466 boxes ha-1 year-1.

From 1999 to 2002 productivity in Costa Rica experienced a fast decline, and sta-bilized at 2107 boxes/ha/yr during 2002. This decrease was due to an inadequate distribution of rainfall: 45% of the total annual precipitation fell from November to December in 1999 and 50% from January to February in 2000. This excessive rainfall was not expected, because of the even rainfall distribution since the second semester of 1997. Plantations did not have adequate superficial drainage; therefore, as a result of soil saturation, and the concomitant anoxia and root death, productivity was low. In addition to inadequate rainfall distribution, the average temperature decreased during this period resulting in fewer flowering plants per hectare (Figure 5). There were fewer than 40 buds ha-1 week-1 for most of the first semester of 2000, which is lower than the minimum accepted value. On average, in the Caribbean coast of Costa Rica the accu-mulated annual precipitation from 1999 to 2002 increased by 6.4% and 17% compared with the previous year. The 17% increase in rainfall in 2002 included two floods that took place in May and December.

1500

2500

3500

4500

5500

1987 1989 1991 1993Year

1995 1997 1999 2001

(mm)

Rainfall

Sarapiqui Pococí SiquirresLimón Matina Talamanca


The decrease of 216 boxes/hectare from 1999 to 2000 and the uncertainty of the glo-bal banana market (Vargas 2000) had the same effect as in 1994, with a decrease in farm maintenance. Drainage, fertilization and nematicide applications were the most affected. Lack of liquidity prevented the continued nematicide rotation that had been recommended by Anderson et al. (1998) and more recently by Moens et al. (2003). The constant use of the same nematicide compound, mainly by independent growers, caused accelerated biodegradation. This, in turn, decreased its efficacy in nematode control and consequently reduced functional root fresh weight at the national level.

The increase in precipitation from 2000 to 2002 generated an increase in the use of the azoxistrobin. At the beginning of 2001, after azoxistrobin had been applied more than 25 times, alternated with other systemic fungicides and protectants, loss of sensitivity became evident in Sarapiquí county (Guzmán 2003). The increase in disease expres-sion at the national level, resulted in fewer leaves at harvest and lower bunch weights throughout the Caribbean coast of Costa Rica.

Finally, plant vigour and the production of the 30,000 hectares planted in the last Banana Development Programme declined slowly between 1988 and 1994. Functional root deterioration in these areas is associated with inappropriate soils and induced acidity caused by the inappropriate use of nitrogen fertilizers in the fertilization band over the years (Figure 6). In addition, it was found that acidity had penetrated deep into the profile, which is related to the plantation’s age, the soil texture and the history of fertilization management practices (Serrano 2003).

It can be concluded that functional root weight was largely responsible for productivity variation. Several abiotic and biotic factors affected functional root weight. Abiotic factors included low and high precipitation, low and high temperatures, poor inter-nal and superficial drainage, accumulated acidity (element toxicity) and the physical

Figure 5. Effect of average weekly temperature on the number of flowering plants per hectare in Pococí county, 1999-2000.

2 Jan 27 Feb 24 Apr 19 Jun 14 Aug 9 Oct 4 Dec 28 Jan 24 Mar 19 May 14 Jul

1999 2000

25

30

35

40

45

50

55

20

21

22

23

24

25

26

27

28

29

30

Bagging Average temperature

Fruiting plants �C


condition of the soil. Biotic factors included nematodes and microorganisms responsi-ble for accelerated nematicide biodegradation. In addition, continuous fluctuations in the international banana market directly influenced the restructuring of farm operation systems; reducing farm maintenance such as drainage systems, fertilization program-mes and nematicide applications. This too affected functional root weights.

ReferencesAnderson J.P.E., H. Haidt & K. Nevermann. 1998. Accelerated microbial degradation of nematicides in soils: problem

and its management. Pp. 568-586 in Memorias XIII Reunión Acorbat (L.H. Hidalgo, ed.). CONABAN, Guayaquil, Ecuador.

Araya M. & A. Cheves. 1998. Selección del tipo de planta para el muestreo de nemátodos en banano (Musa AAA). INFOMUSA 7(1):23-26

Figure 6. Induced accumulated acidity in soils of banana farms located west and east of the Reventazón River in the Caribbean coast of Costa Rica, 1991-2002.

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Col(+)/L

West zoneEast zone

y = 0.242x + 0.7589

R2 = 0.819

P < 0.003y = 0.3136x - 0.1318

R2 = 0.9708

P < 0.0001

0

1

2

3

4

5

1991 1992 1993 1994 1995 1996Year

1997 1998 1999 2000 2001 2002

Col(+)/L

East zone West zone

4,5

4

3,5

3

2,5

2

1,5

`1

0,5

0

The problem of banana root deterioration and its impact on production34 Relatioship between functional root content and banana yield in Costa Rica34 C.A. Gauggel et al. 35

Calvo C. & M. Araya. 2001. Cantidad de raíces de banano en los diez cantones productores de Costa Rica. Corbana 27:47-64.

Davide R.G. 1980. Influence of cultivar, age, soil texture and pH on Meloidogyne incognita and Radopholus similis on banana. Plant Disease 64:571-573.

Durán M.A. 1988. Determinación de la relación de biotipos, resistencia a Fenamiphos, reproducción y patogenicidad de Radopholus similis, en la zona bananera de Izabal, Guatemala. Tesis de Ing. Agr. Ciudad de Guatemala, Facultad de Agronomía, Universidad de San Carlos. 35pp.

Guzmán M. & R. Romero. 1997. Comparación de los fungicidas Azoxistrobina, Propiconazole y Difenoconazole en el control de la Sigatoka negra (Mycosphaerella fijiensis Morelet) en banano (Musa AAA). Corbana 22:49-59.

Guzmán M. 2003. Resistencia a fungicidas en Mycosphaerella fijiensis: Situación actual y perspectivas para el manejo de la Sigatoka negra. Pp. 23-25 in Manejo convencional y alternativo de la Sigatoka negra, nematodos y otras plagas asociadas al cultivo de Musáceas - Resúmenes (G. Rivas, ed.). INIBAP, MUSALAC & FUNDAGRO. Guayaquil, Ecuador.

Moens T., M. Araya, R. Swennen & D. De Waele. 2003. Biodegradación acelerada de nematicidas después de apli-caciones repetidas en una plantación comercial de banano. Pp. 35-36 in Manejo convencional y alternativo de la Sigatoka negra, nematodos y otras plagas asociadas al cultivo de Musáceas - Resúmenes (G. Rivas, ed.). INIBAP, MUSALAC & FUNDAGRO. Guayaquil, Ecuador.

Navarro M. 1995. Rational management of the Nemacur applications (talk). Carmen 2, Siquirres, Bayer, Technical Assistance.

Romero R. & Sutton T. 1997. Sensitivity of Mycosphaerella fijiensis, causal agent of black Sigatoka of banana to propiconazole. Phytopathology 87:96-100.

Sánchez O. & M. Zuñiga. 2001. Informe Estadístico de Exportaciones Bananeras 1970- 2000. Dirección de Estadística y Política Bananera. Costa Rica. CORBANA S.A., Costa Rica. 18pp.

Sánchez O. & M. Zuñiga. 2003. Costa Rica informe anual de estadísticas de exportación de banano. CORBANA S.A., San José, Costa Rica. 54pp.

Serrano E. & D.H. Marín. 1998. Disminución de la productividad bananera en Costa Rica. Corbana 23:85-96.

Serrano E. 2003. Diagnóstico físico-químico del suelo y su relación con el crecimiento del cultivo del banano en fincas con diferentes condiciones edafoclimáticas de la zona Caribe de Costa Rica. Pp. 113-117 In Informe anual 2002, Dirección de Investigaciones (J. Sandoval, ed.). CORBANA, San José Costa Rica.

Vargas R. 2000. CORBANA S.A. y su plan estratégico de investigaciones. Corbana 26:75-80.

Zúñiga M. 1994. Surgimiento del “Acuerdo Marco”. Corbana 19:1-4.

The problem of banana root deterioration and its impact on production34 Relatioship between functional root content and banana yield in Costa Rica34 C.A. Gauggel et al. 35

2

Root anatomy and morphology

Anatomía y morfología de la raíz

C.A. Gauggel et al. 37N. Vazquez M. 37

Anatomy and morphology of monocotyledonous and dicoty-ledonous rootsNelly Vásquez M.1

AbstractIn most plants, the root is the sporophyte’s subterranean part carrying out activities such as anchorage, storage of food products, absorption of water and nutrients and conduction to the remaining plant parts. Roots also produce soil organic matter and provide habitat for mycorrhizae, rhizosphere and rhizoplane organisms. The first root of a plant originates in the embryo and constitutes the primary, main, or tap root. Lateral or secondary roots originate from it. In the monocots, the primary root dies early and the root system is made up of adventitious roots originating from the hypocotyl and sometimes from the stem.

At the root apex of monocots and dicots, three growth areas are recognized: an area of cell division formed by the root meristem; a cell enlargement region, and a maturation zone. The apex is protected by a layer of cells called the root cap, made of tissues with a different structure and function, among which the root cap’s own formation and mucilage production and secretion, are the most important. The secretion of mucilage facilitates root growth through the soil and the root cap controls the root’s gravitropic response.

In general terms, the root has a primary and a secondary structure that facilitate the study of the tissues within it. In primary growth, the following tissues are identified: epidermis (usually unicellular), cortex (storage and axial transportation of water and other materials), endodermis, pericycle and vascular tissue. In roots with primary growth, the xylem and the phloem are in alternate positions, one beside the other, showing a centripetal maturation. In many dicots, secondary growth develops later. It is important to mention that lateral roots have an endogenous origin, probably from the pericycle and endodermis. Adventitious roots, characteristic of the monocots are usually of endogenous origin, though they can sometimes be exogenous.

Finally, roots make connections and relationships with other organisms such as bacteria and mycorrhizae to work together for mutual benefit.

Resumen - Anatomía y morfología de raíces monocotiledóneas y dicotiledóneasEn la mayoría de las plantas la raíz es la parte subterránea del esporófito y realiza funciones de anclaje, almacenamiento de productos para alimentación, absorción de agua y nutrientes. Las raices también tienen funciones de: conducción de agua y nutrientes hacia el resto de partes de la planta, producción de materia orgánica al suelo y provisión de hábitats para micorrizas y organismos de la rizosfera y el rizoplano, etc. La primera raíz de la planta se origina en el embrión y constituye la raíz primaria, principal o pivotante. En monocotiledóneas, la raíz primaria muere pronto y el sistema radical queda formado por raíces adventicias que se originan en el hipocotilo y a veces en el tallo. Este tipo de raíz generalmente crece de manera superficial, lo que las hace aptas para prevenir la erosión.

A nivel del ápice de la raíz tanto de monocotiledóneas como de dicotiledóneas, se reconocen tres áreas de crecimiento: una región de división celular formada por el meristemo radical y la zona cercana de división celular; una región de alargamiento celular y una región de maduración. El crecimiento del ápice está protegido por una capa de células llamado caliptra, constituída por tejidos de diferente estructura y función, entre las que destacan la misma formación de la caliptra,

1 Histology Laboratory, CATIE, Turrialba, Costa Rica, e-mail: [email protected]

The problem of banana root deterioration and its impact on production38 Anatomy and morphology of monocotyledonous and dicotyledonous roots C.A. Gauggel et al. 39N. Vazquez M. 39

la producción y secreción de mucílago que facilita el crecimiento de la raíz a través del suelo, y la gravipercepción de la raíces.

En términos generales, la raíz consta de una estructura primaria y una secundaria que facilitan el estudio de los tejidos que la forman. En el crecimiento primario se identifican tejidos como: epidermis usualmente unicelular, corteza (almacenamiento y transporte lateral de agua y otros materiales), endodermis, periciclo y tejido vascular. En las raíces con crecimiento primario, el xilema y el floema se encuentran en posición alterna, uno al lado del otro y presentan una maduración centrípeta. En muchas dicotiledóneas, se desarrolla posteriormente un crecimiento secundario

Es importante mencionar que las raíces laterales tienen un origen endógeno, posiblemente a partir del periciclo y la endodermis. Las raíces adventicias, típicas de las monocotiledóneas, presentan un origen usualmente endógeno, aunque a veces puede ser exógeno. Por último, las raíces forman conexiones y relaciones con otros organismos, como las bacterias y micorrizas para trabajar juntas por el beneficio mutuo.

IntroductionIn most plants, the root is the part located underground which serves to anchor the plant, store food products, absorb water and nutrients, conduct water and nutrients to the rest of the plant parts. Roots also aid in: producing organic matter for the soil and providing habitats for mycorrhizal and rhizosphere organisms. In addition to these functions, the root synthesizes, accumulates and secretes a large quantity of compounds called exudates, through which it regulates the microbial community in the soil immediately surrounding it. Roots also change the physical and chemical soil properties to favour symbiotic asso-ciations and inhibit the growth of competing plants (Estabrook and Yoder 1998, Walker et al. 2003). Apparently 5 to 21% of all the carbon fixed photosynthetically is transferred to the rhizosphere through exudates (Marschner 1995). The first root of any plant originates in the embryo and constitutes the primary, principal or tap root, which reaches deeper into the soil than any other. Later, lateral or secondary roots grow from the primary root possibly originating from the pericycle or the endodermis.

In monocotyledons, and in some dicotyledons, the primary root dies quickly and the remaining root system is formed by adventitious roots that originate from the hypocotyl and sometimes from the stem. Thus, this adventitious system results in a fibrous root system in which no single root predominates over another. It is important to recognize the different types of roots since each has a different function. The majority of plants produce one or more orders of lateral roots. The different orders vary in thickness, bran-ching pattern, growth rate, capacity to produce secondary growth and structural charac-teristics. These variations can influence their capacity to obtain water and nutrients and to survive under adverse conditions.

Three different growth regions can be identified in both monocotyledon and dicotyledonous roots: an apical cell division region formed by the root meristem; a region of cell lengthe-ning; and a region of cell maturation. The apical zone is composed of a quiescent center and peripheral zone, with very active mitotic cells that form the larger part of the root (Figure 1). Frequently, three possible functions are mentioned for this region (Flores 1999):

• Source of cytokinin

• Model to generate and modify root histological patterns

• Regulate mitotic differentiation and properties of adjacent initial cells


The cell elongation zone measures a few mm in length and, as its name would indicate, it is the area where the cells grow in volume. The maturation zone is where the primary tissues become differentiated and root hairs develop from the trichoblasts. Nevertheless, these regions are not clearly differentiated; however, identifying them makes it easier to study this organ of the plant.

The lateral and the adventitious roots have root meristems similar to that of the primary root. Apical growth is protected by a layer of cells called the root cap (calyptra). It is covered by a layer of highly hydrated mucilage that lubricates the root as it penetrates the soil, protects it from desiccation, favours ion absorption and provides an ideal envi-ronment for the establishment and development of bacteria in the rhizophere. This cel-lular layer also has an important role in sensing gravity (Rougier and Chaboud 1985). In general terms, roots have a primary structure and a secondary structure that facilitate the study of the tissues that compose it.

Primary structure Tissues that can be identified in primary growth are: a generally unicellular epidermis, the cortex (responsible for storing and axial transportation of water and other mate-rials), the endodermis, the pericycle and the vascular tissue or stele.

The epidermisThe main function of the epidermis is water and nutrient absorption, facilitated by root hairs. These root hairs are unicellular, have a short lifespan and are limited to the region of cell maturation. Young roots with live epidermal cells and root hairs are frequently considered those responsible for the greatest nutrient uptake by increasing the surface area of the root (Marschner 1995, Peterson and Farquhar 1996) (Figure 2). The oldest epidermis may be covered by a cuticle and suberin layer. In spite of this, the epidermal cells offer little resistance to the passage of water and nutrients.

The cortex The most space in the primary root body is occupied by the cortex. It is frequently composed of cells from the parenchyma and can have large gas spaces, store starch,

Figure 1. Root apical meristem. Figure 2. Root cap development.


proteins, and fat globules and lacks chloroplasts (Figure 3). It can also contain cellu-lar inclusions such as crystals or secondary metabolites. The latter frequently provide colour and can auto-fluouresce. Many plants accumulate large quantities of dark brown tannins although they could have other coloured pigments. This natural pigmentation of the roots can help distinguish young roots from older roots, which are generally darker (Brundrett, sf.). The parenchyma tissue that forms the cortex also plays a critical role in the later transport of water and other substances via the apoplast and symplast. In monocotyledons, the cortex is maintained throughout the plant’s life and the cells can develop secondary walls and lignify. In dicotyledons, the cortex is shed when secon-dary growth begins.

EndodermisThe endodermis tissue develops within the cortex and is characterized by a more com-pact, generally unicellular layer that has no gas space. It also has Casparian bands in the radial and transverse anticlinal walls (Figure 4). These cells have lignin and suberin deposits making them impermeable to water and ions. Any substances that they allow through to the stele must pass through the protoplasm of the Casparian band cells which is accomplished through the plasmalemma or the plasmodesmata. In many plants, just beyond the endodermis is a parenchymal layer called the pericycle. This layer is meris-tematic and is the source of lateral roots, although it can sometimes lignify. Beyond the pericycle is the vascular cylinder or stele. The roots of some monocotyledons, like corn, have a medulla as well as a vascular cylinder.

Vascular tissue The vascular cylinder of the root is very different from that in the stem. In the primary stem, the xylem and the phloem are found in the same vascular bundle with the phloem generally toward the outside. In primary roots, the xylem and the phloem are found in alternating positions, next to one another. They also have centripetal maturation. In monocotyledons, either one single metaxylem element is in the central position or several poles of xylem and phloem surround a central medulla (Figure 5). This is the typical organization in Musa roots. In many dicotyledons, secondary growth develops later where the cambium and the peridermis play an important role. These roots with secondary growth shed their cortex and endodermis early.

Secondary structureMonocotyledon roots do not have secondary growth. Nevertheless, in some monocoty-ledons there is a primary thickening meristem at the distal end (very near the apex and that diminishes basipetally). Secondary growth typical of the dicotyledons consists of secondary xylem and phloem production from the vascular cambium. Peridermis pro-duction is also seen from the other secondary meristem called the phellogen (Figure 6). The peridermis replaces the epidermis during the root’s secondary growth. Some parts of the peridermis are differentiated like lenticels (spongy tissue that participates in gas exchange). The phellogen produces cork toward the outside and phellodermis toward the inside.


Relationship with other organismsPlants are in constant communication with a large variety of organisms, some of which are beneficial. Among these there are several notable symbiotic associations with bac-teria that fix nitrogen and with mycorrhizal fungi. Other organisms, such as some fungi and bacteria, microbes, viruses and nematodes are damaging to the plant (Estabrook and Yoder 1998). Symbiotic associations between the root and organisms that fix N2 (Rhizobium) are very common in the Fabaceae family (Dart 1977). Bacteria provide nitrates and plants provide carbohydrates.

Mycorrhizae Symbiotic associations between fungi and plant roots are present in 90% of all higher plants. Only a few families lack mycorrhizal associations. Mycorrhizae extend the root absorption area significantly and make P, Cu and Zn available for the plant. Plants with few root hairs are very dependent on mycorrhizae that can provide resistance to heavy metal pollution, adverse pH, drought, high temperatures and salinity, and scarcity of soil micronutrients. The root exerts a regulatory influence on shoot growth. For exam-ple, drought, flooding, mineral nutrient deficiency, salinity or compaction, provoke

Figure 3. Epidermis and bark development. Figure 4. Development of typical endodermis with Casparian bands.

Figure 5. Development of Phleotherm endodermis with conducting cells.

Figure 6. Xylem and phloem development in alternate positions.

The problem of banana root deterioration and its impact on production42 Anatomy and morphology of monocotyledonous and dicotyledonous roots C.A. Gauggel et al. 43G. Blomme et al.

typical reactions such as reduced foliar expansion, closed stomata, epinasty (curving) of the leaves and senescence of the oldest leaves.

ReferencesBrundrett M. CSIRO Forestry and Forest Products: Roots. An introduction to the root structures which influence mycor-

rhizal fungi. http://www.ffp.csiro.au/research/mycorrhiza/root.html.

Dart P. 1977. Infection and development of leguminous nodules. Pp 367-472 in A Treatise on Dinitrogen Fixation. (R.W.F. Hardy & W.S. Silver, eds). Cambridge University Press, UK.

Estabrook E.M., & J.I. Yoder. 1998. Plant-plant communications: rhizosphere signaling between parasitic angiosperms and their hosts. Plant Physiology 116:1-7

Flores E. 1999. La planta: estructura y función. Editorial Tecnológica de Costa Rica.

Marschner H. 1995. Mineral nutrition of higher plants. Academic Press, London.

Peterson R.L. & M.L. Farquhar. 1996. Root hairs: specialized tubular cells extending root surfaces. The Botanical Review 62:1-40.

Rougier M. & A. Chaboud. 1985. Mucilages secreted by roots and their biological function. Israel Journal of Botany 34:129-146.

Walker T.S., H.P. Bais, E. Grotewold & J.M. Vivanco. 2003. Root exudation and rhizosphere biology. Plant Physiology

132:44-51.

The problem of banana root deterioration and its impact on production42 Anatomy and morphology of monocotyledonous and dicotyledonous roots C.A. Gauggel et al. 43G. Blomme et al.

Methodologies for root system assessment in bananas and plantains (Musa spp.)1G. Blomme, K. Teugels2, I. Blanckaert2, G. Sebuwufu3, R. Swennen2 and A. Tenkouano4

AbstractThe root system of field-grown plantain and banana (Musa spp.) plants has been little investigated because whole plant excavation is tedious and time consuming. Four methods for measuring root system size were assessed. A first approach, early screening, determines the entire root system of nursery-grown juvenile plants to forecast root growth in the field of adult plants. Few significant root trait correlations were observed between juvenile and adult plants. Hence root characteristics of adult plants cannot be adequately estimated from the root size of young plants. A second approach was based on regression of root traits on shoot traits. Regression equations attributed at least 90% of the variation in root growth to variation in shoot development, but models were not universal. Indeed shoot-root ratios vary according to the plant developmental stage and environmental conditions and thus need to be assessed at each location. A third method relied on soil core sampling. Root measurements in two soil core samples could estimate the size of the entire root system with at least 80% accuracy. This method is attractive because it only requires 5% of the time needed to excavate the whole root system. In addition, models to estimate root traits of the mat (mother with daughter plants) were significant irrespective of the suckering behaviour of the variety. The fourth approach consisted in assessing electrical capacitance to quantify root biomass as done in other plant species. Musa plant tissue contains a lot of water and thus secures a high electrical conductivity. Very few significant correlations between capacitance values and root system traits were found. In conclusion, root system assessment based on root core sampling is proposed as it estimates well the size of the entire root system in a fast, cheap and non-destructive way.

Resumen - Metodologías para evaluar el sistema radical en bananos y plátanosLa investigación enfocada al sistema radical de plantas de banano y plátano (Musa spp.) desarrolladas en el campo ha sido muy poca, debido a que excavar plantas completas es tedioso y lento. Se evaluaron cuatro métodos para medir el tamaño del sistema radical. El primer método, que es el de evaluación temprana, determina todo el sistema radical de las plantas juveniles desarrolladas en el invernadero para pronosticar el crecimiento de la raíz de plantas adultas en el campo. Se observaron únicamente unas pocas correlaciones significativas para las características de la raíz entre plantas juveniles y adultas. Por lo tanto, las características de la raíz de plantas adultas no pueden ser estimadas adecuadamente por el tamaño de la raíz de las plantas jóvenes. El segundo método se basó en la regresión de las características de la raíz sobre las características de los hijos. Las ecuaciones de regresión atribuyeron al menos un 90% de la variación en crecimiento de la raíz a la variación en el desarrollo del hijo; sin embargo, los modelos no fueron universales. En efecto, los cocientes hijo-

1 International Institute of Tropical Agriculture (IITA), High Rainfall Station, PMB 008 Nchia-Eleme, Rivers State, Nigeria. Present address: INIBAP-ESA, P.O.Box 24384, Kampala, Uganda, e-mail: [email protected]

2 Laboratory of Tropical Crop Improvement, Katholieke Universiteit Leuven (K.U.Leuven), Kasteelpark Arenberg 13, 3001 Leuven, Belgium, e-mail: [email protected]

3 Crop Science Department, Makerere University, P.O.Box 7062, Kampala, Uganda, e-mail: [email protected] Humid Forest Ecoregional Center (Yaoundé), International Institute of Tropical Agriculture, BP 2008 Messa,

Yaoundé, Cameroon, e-mail: [email protected]

The problem of banana root deterioration and its impact on production44 Methodologies for root system assessment in bananas and plantains (Musa spp.) C.A. Gauggel et al. 45G. Blomme et al.

raíz varían de acuerdo con la etapa de desarrollo de la planta y con las condiciones ambientales, por lo que deberán ser evaluadas en cada sitio. El tercer método se basó en el muestreo de suelo con barreno. Las mediciones de raíz en dos muestras de suelo con barreno podrían estimar el tamaño completo del sistema radical con al menos un 80% de exactitud. Este método resulta atractivo ya que se tarda únicamente un 5% del tiempo que llevaría excavar el sistema radical completo. Además, los modelos para estimar las características de la raíz de la unidad de producción (madre con plantas hijas) fueron significativos, sin importar el comportamiento de los hijos de la variedad. El cuarto método consistió en evaluar la capacitancia eléctrica para cuantificar la biomasa de la raíz, como se realiza en otras especies de plantas. El tejido de las plantas de Musa contiene mucha agua, lo que asegura una alta conductividad eléctrica. Se encontraron muy pocas correlaciones significativas entre los valores de capacitancia y las características del sistema radical. En conclusión, se propone utilizar el método de muestreo de raíz con barreno, ya que éste ofrece una buena estimación del tamaño total del sistema radical, de una manera rápida, barata, y sin destrucción.

IntroductionRoots constitute the link between the plant and the soil thereby providing anchorage and guaranteeing nutrient and water uptake. Roots of plantain and banana (Musa spp.) are also sources of plant growth regulators that contribute to lateral shoot (known as suckers) development and thus perennial growth (De Langhe et al. 1983, Martin Prével 1987, Stover and Simmonds 1987, Lahav and Turner 1989, Price 1995).

Research on Musa root systems is remarkably limited despite the paramount problems with nematodes and longevity and has largely been targeted at high value dessert bana-nas (Moreau and Le Bourdelles 1963, Beugnon and Champion 1966, Lassoudière 1978, Avilán et al. 1982). Moreover, few studies have been carried out on field-grown plants because traditional methods of root assessment under field conditions require tedious and time-consuming excavation of whole plants (Box 1996). In banana, two man-days are indeed needed to excavate one complete adult, ready-to-harvest, Musa spp. plant (Blomme 2000). In addition, excavation is destructive and thus complicates a dynamic assessment of root growth. Root measurement techniques that are non-destructive, quick and efficient, and provide insight into root dynamics, are critical for understan-ding environmental effects on root development. In addition, they constitute a good basis for investigating genetic differences among a large number of cultivars.

An alternative to field assessment of root systems of adult plants could be the eva-luation of juvenile plants in the nursery, provided that there is a good correlation between the root system size at the two developmental stages. Swennen et al. (1986) evaluated juvenile Musa spp. plants grown under hydroponic conditions and were able to demonstrate large differences in root development between different genotypes. However, root growth in this artificial medium may differ from that of field-established plants. Furthermore, plants in hydroponics cannot be grown to their adult stage. The same restriction occurs in rhizotrons as described by Lavigne (1987). Hence, it is not known whether root measurements from juvenile plants adequately reflect root charac-teristics of adult plants.

A second alternative to the field assessment of entire root systems could be through indirect determination of root traits from shoot growth. Indeed, root and shoot growth are highly related in many plants such as in cotton (Gossypium hirsutum L.), pea (Pisum sativum L.), carrot (Daucus carota L.), turnip (Brassica rapa L.), palm spe-cies [Roystonea regia (HBK.) O.F. Cook, Coco nucifera L., Syagrus romanzoffiana


(Chamisso) Glassman and Phoenix roebelenii O’Brien] and strawberry (Fragaria xananassa Duch.) (Pearsall 1927, Broschat 1998, Fort and Shaw 1998). Russell (1977) mentioned that nodal root development in the cereals winter wheat (Triticum aestivum L.) and pearl millet (Pennisetum glaucum L.) could be estimated from the number of leaves. Henderson et al. (1983) found that the extent of root branching was very regular for Sitka spruce and could be estimated from the aboveground stem diameter. Smith (1964) reported that root size of Douglas fir, lodgepole pine and other British Columbia tree species could be estimated from aboveground measurements. In banana, Swennen (1984) and Blomme and Ortiz (1996) established positive correlations between root traits and aboveground plant characteristics, while Gousseland (1983) found that the number of cord roots of the ‘Giant Cavendish’ dessert banana can be estimated from the leaf area. This indicates that root growth is linked to shoot growth in banana, but environmental or developmental influences were not investigated.

Root sampling constitutes a third alternative to quickly estimate the total root size. Fort and Shaw (1998) found substantial correspondence between variability for soil core samples and whole-plant root mass of strawberry, indicating that changes in the root system growth can be effectively estimated from soil cores. Soil core samples required no more than 10% of the time to collect and process the entire plant root system. It is not clear whether this approach is applicable to plantain and bananas.

Research in carrot, maize, oats, onion, sunflower and tomato showed that root mass can be indirectly estimated from the plant’s capacitance value (Chloupek 1972, Chloupek 1977, Dalton 1995, Van Beem et al. 1998). As the Musa plant tissue contains a lot of water, thus securing a high electrical conductivity, the potential of this root assessment method was investigated.

Banana and plantain agronomy and breeding stand to gain a lot from improved knowledge of root growth and development. Since excavation of entire root systems is laborious, destructive and provides non-dynamic insights, methods that are fast, easy to execute and that can be repeated over time on a same plant are needed. Therefore, the objectives of this study were to determine (i) the relationships between juvenile and adult root size, (ii) the relationships between root size and shoot traits, (iii) the optimum core sampling method for adequate estimation of the root system size and (iv) the rela-tionship between capacitance values and root system size.

Materials and methodsSite characteristics Experiments were carried out at the High Rainfall station of the International Institute of Tropical Agriculture at Onne (4°42’ N, 7°10’ E, 10 m altitude above sea level) in southeastern Nigeria. The station is located in a degraded rainforest-swamp. Annual rainfall is 2400 mm, unimodal and distributed from February until November. The soil is derived from coastal sediments. It is a deep and freely drained Typic Paleudult/Haplic Acrisol (FAO/ISRIC 1998) of the coarse-loamy, siliceous isohyperthermic family. Nutrient status is poor, except for P, and pH is low (pH 4.3 in 1:1 H2O in the upper 15 cm). Detailed characteristics of the station have been described elsewhere (Ortiz et al. 1997).


Agronomic practicesAll fields had been under eight years of grass fallow before no-tillage planting of Musa spp. All fields were maintained similarly and this included the nematicide Nemacur applica-tion (a.i. fenamiphos) at a rate of 15 g/plant (3 treatments/year). In addition, fertilization was done with muriate of potash (50% K) at a rate of 600g/plant/year, and urea (46% N) at a rate of 300 g/plant/year, split over 6 equal applications during the rainy season. No mulch was applied. Furthermore, the fungicide Bayfidan (a.i. triadimenol) was applied 3 times/yearat a rate of 3.6 ml/plant to reduce black Sigatoka (Mycosphaerella fijiensis Morelet).

Early root screeningThree experiments were carried out. The first and second experiment involved, respec-tively, in vitro micro-propagated and sucker-derived plants of eight polyploid geno-types (Table 1) including the dessert bananas ‘Yangambi km 5’ and ‘Valery’ (AAA group), the plantain ‘Obino l’ewai’ (AAB group), the cooking bananas ‘Cardaba’ and ‘Fougamou’ (ABB group), the plantain-derived hybrids ‘PITA2’ and ‘PITA7’, and the cooking banana hybrid ‘FHIA-03’ (Swennen 1990b, Daniells et al. 2001). The third experiment involved in vitro-derived plants of 8 diploid accessions (Table 1), namely, ‘Niyarma yik’, ‘Calcutta 4’, ‘Pahang’, ‘Pisang jari buaya’, ‘Pisang madu’, ‘Tjau laga-da’, ‘Kisubi’ and ‘Pisang Berlin’. Micropropagated plants were produced following standard shoot-tip culture techniques (Vuylsteke 1989 and 1998). The plantlets were acclimatized for 6 weeks in a greenhouse nursery (Vuylsteke 1998, Vuylsteke and Talengera 1998), before field transplantation in May and June 1996, for the first and third experiment, respectively. In the second experiment, suckers were prepared as recommended by Swennen (1990a) and field planted in June 1996.

The experimental layout was a split plot within a randomized complete block design with two replications of two plants per genotype. Main plot treatments consisted of different times of observation. In the first and third experiment, data were collected from 6-week-old nursery plants, while field measurements were done on 12, 16 and 20-week-old plants, and at bunch emergence. In the second experiment, measurements were carried out in the field on plants aged 6, 12, 16 and 20 weeks, and at bunch emer-gence. Subplot treatments consisted of genotypes. Plant spacing in the field was 2 m x 2 m, except for plants evaluated at bunch emergence, which were spaced 4 m x 4 m to avoid root entanglement between neighbouring plants. The fields were irrigated during the dry season at a rate of 100 mm per month.

Data collection was carried out on the following corm and root characteristics: corm fresh weight (CW, g), corm height (CH, cm), widest width of the corm (WW, cm), num-ber of suckers (NS) on the corm, number of adventitious or cord roots (NR), cord root length (LR, cm), the average diameter of cord roots at their base (AD, mm), and root dry weight (DR, g). The cord root length was measured using the line intersect method (Tennant 1975), while the diameter of the root was measured with Vernier calipers.

Estimating root traits from shoot traits Twenty-seven genotypes representing the various Musa genome and ploidy groups were assessed in this experiment (Table 1). The planting material was obtained through meristem culture using the methods of Vuylsteke (1989 and 1998). Field transplanting


Table 1. Name, genome/parentage, ploidy level, type and suckering behavior of genoty-pes evaluated for rooting. Name Genome Ploidy Type Suckering ES* MR SC CM paren- level behaviour# tage

‘Niyarma yik’ AA 2 Musa acuminata ssp. banksii Inhibited 3 X

‘Calcutta 4’ AA 2 Musa acuminata ssp. burmannica Non-regulated 3 X 1 X

‘Pahang’ AA 2 Musa acuminata ssp. malaccensis Non-regulated 3 X

‘Pisang jari buaya’ AA 2 Musa acuminata ssp. microcarpa Regulated 3 X

‘Pisang madu’ AA 2 Musa acuminata ssp. microcarpa Regulated 3 X

‘Tjau lagada’ AA 2 Musa acuminata ssp. microcarpa Regulated 3 X

‘Kisubi’ AB 2 Dessert banana Regulated 3 X

‘Pisang Berlin’ 2 Diploid indeterminate group Regulated 3 X

TMB2x 9128-3 AA x AA 2 Hybrid (Tjau lagada x Pisang lilin) Non-regulated X 2

TMB2x 5265-1 AA x AA 2 Hybrid (Tjau lagada x Calcutta 4) Non-regulated X

TMP2x 1297-3 AAB x AA 2 Plantain hybrid (Agbagba french reversion x Calcutta 4) Non-regulated X

TMP2x 2829-62 AAB x AA 2 Plantain hybrid (Bobby tannap x Calcutta 4) Non-regulated 2

‘Pisang M. hijau’ AAA 3 Dessert banana Regulated X

‘Yangambi km5’ AAA 3 Dessert banana Non-regulated 1&2 X X

‘Valery’ AAA 3 Dessert banana Regulated 1&2 X X

‘Rajapuri’ India AAB 3 Dessert banana Regulated X

‘Obino l’ewai’ AAB 3 Plantain Inhibited 1&2 X X

‘Bobby tannap’ AAB 3 Plantain Inhibited X

‘Mbi egome’ AAB 3 Plantain Regulated 1 X

‘Muracho’ AAB 3 Starchy banana Regulated X

‘Mysore’ AAB 3 Dessert banana Regulated X

‘Pisang awak’ ABB 3 Cooking banana Regulated X

‘Foulah 4’ ABB 3 Cooking banana Regulated X

‘Cardaba’ ABB 3 Cooking banana Regulated 1&2 X X

‘Fougamou’ ABB 3 Cooking banana Regulated X X

IC 2 AAAA 4 Dessert banana Regulated X

TMP4x 1621-1 AAB x AA 4 Plantain hybrid (Obino l’ewai x Calcutta 4) Regulated X

TMP4x 548-9 AAB x AA 4 Plantain hybrid (Obino l’ewai x Calcutta 4) Regulated 1&2 X X

TMP4x 5511-2 AAB x AA 4 Plantain hybrid (Obino l’ewai x Calcutta 4) Inhibited X

TMP4x 1658-4 AAB x AA 4 Plantain hybrid (Obino l’ewai x Pisang lilin) Regulated 1&2 X

FHIA-03 ABB x AA 4 Cooking banana hybrid (SH-3386 x SH-3320) Regulated 1&2 X* CM: Capacitance Method, ES: Early Screening method, MR: Multiple Regression method, SC: Soil Core method. Within the methods, numbers show the experiment number in which the method was used. #: Suckering behaviour is inhibited when suckers do not grow fast until flowering of the parent plant; is regulated if 1-3 suckers grow very well before flowering; is non-regulated if almost all suckers grow well before flowering.


was done in June 1996 at a spacing of 2 m x 2 m. The experimental design was a ran-domized complete block with two replications of two plants per genotype. Shoot and root traits were assessed during mid-vegetative growth (i.e. 20-week-old plants). Shoot growth characteristics included plant height (PH, cm), circumference of the pseudos-tem at soil level (PC, cm), height of the tallest sucker (LS, cm), number of leaves (NL) and leaf area (LA, cm2). Leaf length and maximum leaf width were measured and LA was calculated according to Obiefuna and Ndubizu (1979). The corm and roots were completely dug out and assessed as described above. Total cord root length (TL, cm) of the mat (i.e. main plant and suckers) and total root dry weight of the mat (TD, g) were also measured. The tallest sucker was separated from the main plant and the same characteristics as above were measured.

Core sampling In the first experiment, the small French plantain ‘Mbi egome’ (Swennen 1990b) and the wild diploid banana ‘Calcutta 4’ were assessed (Table 1). ‘Calcutta 4’ has a non-regulated suckering (all suckers grow simultaneously), while ‘Mbi egome’ has a regula-ted suckering behaviour (one or two suckers grow vigorously). Sucker-derived plants of ‘Mbi egome’ and in vitro-derived plants of ‘Calcutta 4’ were field established in August 1997. Suckers were prepared and planted according to Swennen (1990a). In vitro-deri-ved plants of Calcutta 4 were grown in polybags for 6 weeks in the greenhouse before field planting. Plant spacing was 4 m x 4 m. In the second experiment, 30 progenies from a cross between two diploid hybrids, ‘TMB2 x 2829-62’ and ‘TMB2 x 9128-3’ were assessed (Table 1). Micro-propagated plants were obtained, as in experiment one, and were established in the field during August 1996 at a spacing of 3 m x 2 m. In both experiments, treatments were completely randomized.

Sixty-week-old mats were assessed in experiment one. Eight soil core samples were taken at 15 cm from the plant base. The first sample was taken near the biggest sucker and where the future axial sucker would emerge. Subsequent samples were taken cloc-kwise at 45° intervals. Soil cores had 25 cm diameter and a 80 cm height. Soil cores were taken with a metal cylinder. Samples were washed to free roots from soil and the following characteristics were measured for each sample: number of adventitious or cord roots (NR), root dry weight (DR, g) and cord root length (LR, cm). The same characteristics were measured for the entire plant, which was excavated after core sam-pling. In the second experiment, root assessment was carried out as in experiment one but in July 1999 on 3-year-old mats. Three soil cores were taken per plant including the position next to the tallest sucker and at 90° and 180° clockwise from the tallest sucker.

Capacitance measurements Measurements were carried out on in vitro and sucker-derived plants of different ages (Table 1). Variability in capacitance values was assessed according to the position of the electrodes (Figure 1). The soil-based positive electrode was placed at different soil depths and distances from the plant. The negative electrode was inserted in the pseu-dostem at variable heights and depths. Also the influence of corm size, soil temperature and water content on capacitance was investigated.


Statistical analysis Statistical analysis was carried out using the SAS statistical package (SAS 1989). For the early screening experiments, simple Pearson correlation coefficients between identical root growth characteristics of plants at different ages across genotypes were calculated. Simple correlation and multiple regression analyses were used to estimate the relationships between aerial growth and root system characteristics. Regression was carried out using stepwise selection with root characteristics as dependent variables and shoot traits as independent variables. Ploidy level was also included as an independent variable in the regression analysis.

Data from the core sampling experiments were subjected to ANOVA to determine the effects of plant and sampling location on soil core root characteristics. Root charac-teristics of the core samples were regressed on whole plant root traits. Data from the capacitance experiments were subjected to ANOVA to determine the effects of the position of the electrodes on the capacitance values. To find relationships between the capacitance values and the root parameters, scatter plots and linear correlation analyses (Proc CORR in SAS) were carried out on the complete data set but also according to the type of planting material and the age of the plants.

Results and discussionEarly root screening Few significant correlations were observed for corm and root traits between the diffe-rent growth stages of diploid, triploid and tetraploid genotypes in all three experiments (Tables 2 and 3). Significant positive correlations were mostly observed for in vitro-derived plants. There were very few correlations for the traits considered in all three experiments between plants at 6 weeks after planting and plants in the mid-vegetative stage (i.e. 20 weeks old) or plants at bunch emergence. An increased difference in plant age reduced the number of significant correlations (Tables 2 and 3). Absence of significant correlations with sucker-derived plants might be attributed to differences in

SD

PSHD

PSD meter

-

+

A B

Figure 1. (A) Traits defining the position of both electrodes (PSH: Pseudostem Height, PSD: Pseudostem Depth, D: Distance from the pseudostem, SD: Soil Depth); (B) Measuring the capacitance; the arrows indicate the electrodes.


physiological age of the suckers at planting, which influenced subsequent growth. In addition, large genotypic variation exists for vegetative growth cycle and plant size at bunch emergence (Swennen and Vuylsteke 1987), which may explain the absence of correlations between root growth during the early vegetative stage and early reproduc-tive stage in both propagule types. We conclude that the root characteristics of adult plants cannot be estimated from the root size of juvenile plants irrespective of the type of planting material.

Estimating root traits from shoot traits Significant correlation coefficients between aerial and root system characteristics were found during vegetative development (Table 4) confirming earlier reports (Beugnon and Champion 1966, Gousseland 1983, Swennen 1984, Lavigne 1987, Blomme and

Table 2. Correlation coefficients between identical growth characteristics at different ages for in vitro and sucker-derived triploid and tetraploid plants. Trait #

in vitro CW CH WW DR NR LR AD TD TL LS NS

6,12 (1) na na na 0.34 0.22 0.38 0.39 0.34 0.37 na na

6,16 na na na 0.22 0.61 0.12 0.4 0.08 -0.02 na na

6,20 na na na 0.41 0.13 0.20 0.05 0.22 0.1 na na

6,Fl (2) na na na 0.75* 0.57 0.30 0.03 0.18 -0.3 na na

12,16 0.17 0.15 0.49 0.71* 0.26 0.47 0.56 0.66 0.5 0.58 0.47

12,20 0.27 0.52 0.34 0.87** -0.03 0.60 0.46 0.91** 0.81* 0.64 0.1

12,Fl -0.32 -0.33 -0.18 0.07 -0.18 -0.45 0.24 0.12 -0.22 -0.03 0.27

16,20 -0.18 -0.03 -0.01 0.62 0.35 0.58 0.73* 0.70 0.6 0.98*** 0.89**

16,Fl 0.22 0.34 0.03 -0.05 0.29 0.18 0.06 0.33 0.42 0.45 0.87**

20,Fl 0.07 -0.07 -0.22 0.37 0.39 -0.16 0.26 0.41 -0.07 0.42 0.87*

Sucker- CW CH WW DR NR LR AD TD TLderived

6,12 -0.10 -0.30 0.03 -0.18 -0.06 0.07 0.14 -0.18 0.07

6,16 0.44 0.60 0.11 0.46 0.55 0.40 -0.23 0.46 0.40

6,20 -0.22 -0.52 -0.69 0.32 -0.10 0.20 0.39 0.32 0.20

6,Fl 0.07 0.39 -0.20 -0.40 -0.09 -0.60 0.15 -0.12 -0.37

12,16 -0.30 -0.69 -0.10 0.22 0.17 0.41 -0.22 0.22 0.41

12,20 0.40 0.48 -0.53 -0.47 -0.20 -0.20 -0.31 -0.47 -0.20

12,Fl 0.36 0.13 0.29 0.29 -0.65 -0.18 -0.50 0.42 0.08

16,20 0.03 -0.19 -0.10 0.64 0.10 0.68 0.15 0.64 0.68

16,Fl -0.35 -0.21 -0.67 0.35 0.13 -0.78* 0.01 0.79* 0.21

20,Fl 0.33 -0.06 0.09 0.33 0.04 -0.32 0.67 0.52 0.32#: AD: average basal cord root diameter (mm), CH: corm height (cm), CW: corm fresh weight (g), DR: root dry weight (g), LR: cord root length (cm), LS: height of the tallest sucker (cm), NR: number of cord roots, NS: number of suckers, TD: total root dry weight of the mat (g), TL: total length of the cord roots of the mat (cm), WW: corm widest width (cm)*, **, *** Significant at P<0.05, 0.01 and 0.001, respectively(1): 6,12: correlation between plants aged 6 and 12 weeks, (2) Fl: at flower emergencena: not applicable.


Ortiz 1996). Similar relationships between root and shoot traits were also found for the East African highland bananas (AAA-EA group) (Sebuwufu et al. 2004). Furthermore, regression analysis produced several equations (Table 5) that attributed at least 90% of the variation in root growth to variation in shoot development. The best shoot indica-tors of root growth were leaf area, pseudostem circumference and height of the tallest sucker. Swennen (1984) already demonstrated that ratooning can be improved by increasing rooting of the parent plant.

These regression models suggest that a reduced leaf area, as caused by black Sigatoka, will adversely affect root development. Hence increased (photosynthetically active) leaf area (from optimal plant spacing, thereby reduced shading) and improved nutrient

Table 3. Correlation coefficients between identical growth characteristics at different ages for in vitro-derived diploid plants. Trait #

in vitro CW CH WW DR NR LR AD TD TL LS NS

6,12 (1) na na na -0.48 -0.37 -0.25 0.79* -0.48 -0.25 na na

6,16 na na na -0.09 -0.27 -0.18 0.33 -0.22 -0.25 na na

6,20 na na na 0.38 0.56 0.52 0.11 0.36 0.46 na na

6,Fl (2) na na na -0.03 -0.76* -0.42 0.05 -0.26 -0.40 na na

12,16 0.61 0.53 0.65 0.46 0.64 0.80* 0.68 0.62 0.85* -0.24 0.11

12,20 0.25 0.66 -0.01 0.13 0.22 0.36 -0.42 0.31 0.45 -0.67 0.48

12,Fl 0.01 -0.09 0.17 0.17 -0.20 -0.35 0.40 0.37 -0.16 -0.06 0.91**

16,20 0.05 0.76* -0.22 0.41 0.36 0.73* -0.13 0.47 0.80* 0.77* 0.13

16,Fl -0.12 -0.03 0.01 0.37 0.03 0.09 0.06 0.15 0.14 0.43 0.02

20,Fl 0.40 -0.43 0.47 0.19 0.04 0.10 -0.15 0.44 0.27 0.48 0.44# AD: average basal cord root diameter (mm), CH: corm height (cm), CW: corm fresh weight (g), DR: root dry weight (g), LR: cord root length (cm), LS: height of the tallest sucker (cm), NR: number of cord roots, NS: number of suckers, TD: total root dry weight of the mat (g), TL: total length of the cord roots of the mat (cm), WW: corm widest width (cm)*, **, *** Significant at P<0.05, 0.01 and 0.001, respectively(1): 6,12: correlation between plants aged 6 and 12 weeks, (2) Fl: at flower emergencena: not applicable.

Table 4. Correlation coefficients (P<0.05) between aerial growth and root system charac-teristics at 20 weeks after planting.Trait # LA PH PC LS

DR 0.72*** 0.65*** 0.65*** -0.09

NR 0.46* 0.41* 0.29 0.16

LR 0.64*** 0.54** 0.46* 0.08

AD 0.47* 0.51** 0.70*** -0.38*

TD 0.65*** 0.53** 0.38 0.2

TL 0.41* 0.25 0.01 0.49*#: AD: average basal cord root diameter (mm), DR: root dry weight (g), LA: leaf area (cm2), LR: cord root length (cm), LS: height of the tallest sucker (cm), NR: number of cord roots, PC: plant circumference (cm), PH: plant height (cm), TD: total root dry weight of the mat (g), TL: total length of the cord roots of the mat (cm)*, **, *** Significant at P<0.05, 0.01 and 0.001, respectively.


status of the plant would stimulate root development. Indeed roots are sinks for assi-milates that need to come from functional leaves. The pseudostem is made up of leaf sheaths and hence reflects the number of leaves and plant vigour. Plant pseudostem circumference thus reflects shoot growth and is an important determinant of root vigour in the regression models.

The size of the tallest sucker reflected positively the extent of the mat root system. Most suckers observed on 20-week-old plants were peepers (i.e. small sucker with scale leaves) or sword suckers (i.e. larger sucker with lanceolate leaves). The latter suckers have their own root system, confirming observations by Robin and Champion (1962) and Beugnon and Champion (1966). The observed positive effect of ploidy on cord root diameter confirms earlier observations by Monnet and Charpentier (1965).

Shoot-to-root ratios depend on the developmental stage of a plant (Brouwer 1966). In banana, Gousseland (1983) estimated cord root number from leaf area and reported an effect of plant developmental phase on the accuracy of the regression model. He reported that the number of cord roots is underestimated during the early vegetative phase. Shoot-to-root ratios also vary across environments (Brouwer and De Wit 1969, Squire 1993, Martinez Garnica 1997, McMichael and Burke 1998, Blomme 2000). Refinement of the regression equations is needed by taking account of the plant deve-lopmental stage and environmental conditions. In conclusion Musa root system deve-lopment can be estimated from easily measurable above ground characteristics. This provides a fast and non-destructive assessment of root development.

Core samplingMost roots of the assessed mats were observed within a 60 cm radius from the plant and up to 70 cm depth. For the first experiment, data on root characteristics obtained through soil core sampling involved about 1.1 to 2.6% of the total root system, depen-ding on the root trait under consideration (Table 6). For ‘Calcutta 4’, a significant inter-plant effect was observed for the dry weight, number and length of the roots present in soil cores (Table 7). This can be explained by the high variability in mat root system size (Table 6). However, no significant differences were observed for sampling

Table 5. Regression models to predict root system characteristics at 20 weeks after planting using aerial growth characteristics and ploidy level as independent variables. Trait # (Independent variables)

Trait # LA PC LS PL R2

DR 0.00163*** 0.596** 0.93

NR 0.00146*** 1.255*** 0.93

LR 0.0667*** 23.47** 0.94

AD 0.0938*** 0.681*** 0.97

TD 0.00207*** 0.426 0.171* 0.93

TL 0.0995*** 14.69*** 0.92#: AD: average basal cord root diameter (mm), DR: root dry weight (g), LA: leaf area (cm2), LR: cord root length (cm), LS: height of the tallest sucker (cm), NR: number of cord roots, PC: plant circumference (cm), PL: ploidy level, TD: total root dry weight of the mat (g), TL: total length of the cord roots of the mat (cm)^: independent variables*, **, *** Significant at P<0.05, 0.01 and 0.001, respectively.


location within a plant (Table 7). This could be due to the high number of well-develo-ped suckers uniformly spaced around the main plant. Significant inter-plant differences in root characteristics were also observed for ‘Mbi Egome’, but the effects of sampling location were even more important (Table 7), which may be explained by the regulated suckering behavior (i.e. development of 1-3 large suckers on the plant) of this cultivar causing sucker root systems to be less uniformly distributed around the main plant.

Root dry weight, number and length of the cord roots in the core samples were regressed on the corresponding characteristics from the mat. All models were signifi-cant at p<0.01, with R2 ≥ 0.58, 0.73 and 0.81 for regressions based on 1 or 2 (Table 8), or 3 core samples (data not shown), respectively. Sampling locations nearest to the tallest sucker did not increase R2 values (Table 8). Data from the second experiment also revealed large differences attributable to variation in suckering behavior among the 30 genotypes assessed. However, this did not appear to affect the regression of core samples on whole plants for root size estimation. The lowest R2 values were obtained with single core samples and were respectively 0.71, 0.81 and 0.78 for root dry weight,

Table 7. Mean squares and significance for different soil core root characteristics for 60 weeks old ‘Calcutta 4’ and ‘Mbi Egome’ mats. Trait #

Genotype Source of variation df DR NR LR

Calcutta 4 Plant 10 16.0*** 141.0*** 55,998***

Sampling location 7 2.17 25.92 7661

Residual 70 2.70 28.30 6981

Mbi egome Plant 11 15.31*** 34.33 25,642***

Sampling location 7 19.65*** 52.78* 23,683**

Residual 77 3.61 20.89 7,244#: DR: root dry weight (g), NR: number of cord roots, LR: cord root length (cm)*, **, *** Significant at P<0.05, 0.01 and 0.001, respectively.

Table 6. Mean and coefficient of variation for different root (whole mat and soil core) characteristics for 60 weeks old ‘Calcutta 4’ and ‘Mbi egome’ mats. Genotype

Calcutta 4 Mbi egome

Trait # Mean CV (%) Mean CV (%)

DR Whole mat 262.9 37 336.0 43

Soil core 2.9 71 4.4 56

Soil core/Whole mat (%) 1.1 66 1.4 52

NR Whole mat 491.0 22 465.5 22

Soil core 13.0 49 11.5 44

Soil core/Whole mat (%) 2.7 44 2.5 44

LR Whole mat 10809.2 31 10200.0 26

Soil core 238.1 47 259.5 40

Soil core/Whole mat (%) 2.2 36 2.6 37#: DR: root dry weight (g), NR: number of cord roots, LR: cord root length (cm).


number of cord roots and cord root length (Table 9). At least 80% of the variation in mat root traits could be explained by taking two core samples, while 85% could be explained with three core samples.

Table 9. R2 values for root dry weight (DR), cord root number (NR) and cord root length (LR), of regressions between single, double and triple soil core samples and the whole mat samples for 30 progenies from a cross between ‘TMPx 2829-62’ and ‘TMPx 9128-3’. Sample No. (1)Trait # 1 2 3 1,2 1,3 2,3 1,2,3

DR 0.78 0.80 0.71 0.86 0.81 0.80 0.85

NR 0.86 0.86 0.81 0.91 0.87 0.89 0.90

LR 0.85 0.86 0.78 0.91 0.85 0.88 0.89#: DR: root dry weight (g), NR: number of cord roots, LR: cord root length (cm)(1) Sample 1 corresponds to site of the future axial sucker; the following samples are taken clockwise around the parent plant at 45 degree intervals.

Table 8. R2 values for root dry weight, cord root number and cord root length, of regres-sions between one (A)/two (B) soil core sample(s) and whole mat samples for 60 weeks old ‘Calcutta 4’ and ‘Mbi egome’ mats. AGenotype Trait # Sample No. (1) 1 2 3 4 5 6 7 8

Calcutta 4 DR 0.73 0.79 0.81 0.80 0.88 0.58 0.78 0.92 NR 0.90 0.91 0.86 0.78 0.93 0.83 0.80 0.86 LR 0.94 0.90 0.92 0.83 0.95 0.88 0.90 0.95Mbi egome DR 0.73 0.75 0.88 0.68 0.79 0.79 0.94 0.87 NR 0.79 0.86 0.93 0.73 0.84 0.95 0.98 0.83 LR 0.89 0.90 0.90 0.84 0.91 0.96 0.98 0.91

B Sample No. (*) 1,2 1,3 1,4 1,5 1,6 1,7 1,8 2,3 2,4 2,5

Calcutta 4 DR 0.82 0.85 0.80 0.85 0.82 0.85 0.88 0.86 0.87 0.90 NR 0.95 0.94 0.88 0.94 0.95 0.91 0.94 0.89 0.92 0.96 LR 0.96 0.97 0.93 0.97 0.97 0.96 0.97 0.92 0.93 0.97Mbi egome DR 0.82 0.90 0.79 0.86 0.80 0.95 0.91 0.87 0.80 0.79 NR 0.90 0.94 0.85 0.89 0.90 0.97 0.92 0.96 0.89 0.87 LR 0.95 0.96 0.92 0.95 0.94 0.99 0.97 0.94 0.92 0.93

2,6 2,7 2,8 3,4 3,5 3,6 3,7 3,8 4,5

Calcutta 4 DR 0.81 0.83 0.89 0.84 0.91 0.89 0.88 0.94 0.91 NR 0.94 0.90 0.92 0.87 0.94 0.92 0.88 0.90 0.90 LR 0.95 0.93 0.96 0.92 0.96 0.96 0.95 0.97 0.94Mbi egome DR 0.83 0.93 0.90 0.88 0.88 0.89 0.94 0.90 0.80 NR 0.95 0.98 0.93 0.92 0.96 0.98 0.97 0.91 0.88 LR 0.96 0.97 0.98 0.94 0.95 0.97 0.96 0.94 0.94

4,6 4,7 4,8 5,6 5,7 5,8 6,7 6,8 7,8

Calcutta 4 DR 0.87 0.89 0.96 0.79 0.87 0.93 0.73 0.84 0.92 NR 0.91 0.88 0.93 0.92 0.92 0.94 0.87 0.92 0.87 LR 0.96 0.96 0.97 0.94 0.95 0.98 0.91 0.96 0.95Mbi egome DR 0.79 0.93 0.84 0.84 0.92 0.88 0.93 0.88 0.93 NR 0.90 0.94 0.85 0.94 0.97 0.92 0.99 0.95 0.94 LR 0.94 0.97 0.92 0.96 0.97 0.98 0.98 0.97 0.97


The results from both experiments indicate that mat root system characteristics can be adequately estimated from two soil core samples. Moreover, root samples collected from two soil cores requires only 5% of the time needed to excavate and measure the entire root system of an adult Musa plant.

Capacitance measurementsDistance of the soil-based electrode from the plant had no significant effect on capa-citance readings (Table 10). In contrast, soil depth, pseudostem height and insertion depth, soil temperature and water content had a significant effect on capacitance. Very few significant correlations between capacitance values and root system traits were found. The specific morphology of the banana plant might be the cause. For example, the enclasping leaf sheaths with their numerous air spaces that make up the pseudostem and the underground corm (i.e. the real stem) with many suckers, might influence the capacitance readings. Therefore root system traits of juvenile and adult field-grown Musa spp. plants cannot yet be estimated through capacitance measurements.

In conclusion, root systems can best be assessed from shoot measurement and root core sampling as both methods are fast, easy to execute and can be repeated over time on the same plant. Hence a dynamic picture on root development can be developed. Both methods however are location specific.

Table 10. Mean square and significance tests for the capacitance value.Source of variation # df Mean square for capacitance

PSH 1 0.325***

PSD 2 0.784***

D 2 0.0057

SD 3 0.590***#: PSH: pseudostem height, PSD: pseudostem depth, D: distance from the pseudostem, SD: soil depth. *, **, *** Significant at P<0.05, 0.01 and 0.001, respectively.

AcknowledgementsFinancial support by the Flemish Association for Development Cooperation and Technical Assistance (VVOB: Vlaamse Vereniging voor Ontwikkelingssamenwerking en Technische Bijstand) and the Directorate General for Development Cooperation (DGDC, Belgium) is gratefully acknowledged.

ReferencesAvilán R.L., R. Meneses & R.E. Sucre. 1982. Distribución radical del banano bajo diferentes sistema de manejo de

suelos. Fruits 37:103-109.

Beugnon M. & J. Champion. 1966. Etude sur les racines du bananier. Fruits 21:309-327.

Blomme G. & R. Ortiz. 1996. Preliminary assessment of root systems. Musa. First International Conference on Banana and Plantains for Africa. MusAfrica 10:16 (abstract).

Blomme G. 2000. The interdependence of root and shoot development in banana (Musa spp.) under field conditions and the influence of different biophysical factors on this relationship. PhD thesis No. 421. KULeuven. Faculty of Agriculture and Applied Biological Sciences. Belgium. 183pp.


Box J.E. Jr. 1996. Modern methods of root investigation. Pp 193-237 in Plant roots: The hidden half (2nd ed.) (Y. Waisel, A. Eshel & U. Kafkafi eds). Marcel Dekker, New York.

Broschat T.K. 1998. Root and shoot growth patterns in four palm species and their relationships with air and soil tem-peratures. HortScience 33:995-998.

Brouwer R. 1966. Root growth of grasses and cereals. Pp. 153-166 in The growth of cereals and grasses (F.L. Milthorpe & J.D. Ivins, eds.). Butterworths, London.

Brouwer R. & C.T. De Wit. 1969. A simulation model of plant growth with special attention to root growth and its consequences. Pp. 224-244 in Root Growth (W.J. Whittington, ed.). Butterworths, London.

Chloupek O. 1972. The relationship between electrical capacitance and some other parameters of plant roots. Biologia Plantarum 14:227-230.

Chloupek O. 1977. Evaluation of the size of a plant’s root system using its electrical capacitance. Plant and Soil 48:525-532.

Dalton F.N. 1995. In-situ root extent measurements by electrical capacitance methods. Plant and Soil 173:157-165.

Daniells J., C. Jenny, D. Karamura & K. Tomekpe. 2001. Musalogue: a catalogue of Musa germplasm. Diversity in the genus Musa (E. Arnaud and S. Sharrock, compil.). INIBAP, Montpellier, France. 213pp. (www.inibap.org/publications/musalogue2.pdf).

De Langhe E., R. Swennen & G.F. Wilson. 1983. Aspects hormonaux du rejetonnage des bananiers plantains. Fruits 38:318-325.

FAO/ISRIC & ISSS. 1998. World reference base for soil resources. World Soil Resources Reports 84. FAO, Rome, Italy. 88pp.

Fort S.B. & D.V. Shaw. 1998. Phenotypic correlations between root and shoot traits of strawberry in fumigated and non-fumigated soils. HortScience 33:222-224.

Gousseland J. 1983. Etude de l’enracinement et de l’émission racinaire du bananier ‘Giant Cavendish’ (Musa acuminata AAA, sous-groupe Cavendish) dans les andosols de la Guadeloupe. Fruits 38:611-623.

Henderson R., E.D. Ford, E. Renshaw & J.D. Deans. 1983. Morphology of the structural root system of Sitka spruce 1. Analysis and quantitative description. Forestry 56:121-135.

Lahav E. & D.W. Turner. 1989. Banana Nutrition. International Potash Institute, Bern, Switzerland.

Lassoudière A. 1978. Quelques aspects de la croissance et du développement du bananier ‘Poyo’ en Côte d’Ivoire. Le système radical. Fruits 33:314-338.

Lavigne C. 1987. Contribution à l’étude du système racinaire du bananier. Mise au point de rhizotrons et premiers résultats. Fruits 42:265-271.

Martinez Garnica A. 1997. Mineral nutrient deficiency in plantain: Symptoms and disorders under experimental and field conditions. Pp. 47-63 in Hohenheim Tropical Agricultural Series 4. Margraf Verlag.

Martin-Prével P. 1987. Banana. Pp. 637-670 in Plant Analysis as a Guide to the Nutrient Requirements of Temperate and Tropical Crops (P. Martin-Prével, J. Gagnard & P. Gautier, eds). Lavoisier. Paris.

McMichael B.L. & J.J. Burke. 1998. Soil temperature and root growth. HortScience 33:947-950.

Monnet J. & J.M. Charpentier. 1965. Le diamètre des racines adventives primaires des bananiers en fonction de leur degré de polyploïdie. Fruits 20:171-173.

Moreau B. & J. Le Bourdelles. 1963. Etude du système racinaire du bananier ‘Gros Michel’ en Equateur. Fruits 18:71-74.

Obiefuna J.C. & T.O.C. Ndubizu. 1979. Estimating leaf area of plantain. Scientia Horticulturae 11:31-36.

Ortiz R., P.D. Austin & D. Vuylsteke. 1997. IITA high rainfall station: Twenty years of research for sustainable agricul-ture in the West African Humid Forest. HortScience 32:969-972.

Pearsall W.H. 1927. Growth studies. VI. On the relative size of plant organs. Annals of Botany 41:449-556.

Price N.S. 1995. Banana morphology 1: roots and rhizomes. Pp.190-205 in Bananas and Plantains (S. Gowen, ed.). Chapman and Hall, London.

Robin J. & J. Champion. 1962. Etudes des émissions des racines de la variété du bananier Poyo. Fruits 17:93-94.

Russell R.S. 1977. Plant root systems: Their function and interaction with the soil. McGraw-Hill Book Company, London.


SAS Institute 1989. SAS/STAT user’s guide, version 6, 4th edition, volume 1. Cary, N.C.: SAS Institute Inc.

Sebuwufu G., P.R. Rubaihayo & G. Blomme. 2004. Variability in the root system of East African banana genotypes in Conservation through utilization of banana and plantain in the Great Lakes Region of East Africa (E. Karamura, D.A. Karamura, S. Sharrock and E. Frison, eds). African Crops Science Journal (Special issue) 12(1):85-93.

Smith J.H.G. 1964. Root spread can be estimated from crown width of Douglas fir, lodgepole pine and other British Columbia tree species. Forestry Chronicle 40:456-473.

Squire G.R. 1993. The Physiology of Tropical Crop Production. CAB International. Wallingford. Pp.33-70.

Stover R.H. & N.W. Simmonds. 1987. Bananas, 3rd ed. Longman, Harlow. 468pp.

Swennen R. 1984. A physiological study of the suckering behaviour in plantain (Musa cv. AAB). PhD thesis, Dissertationes de Agricultura No. 132, KULeuven, Faculty of Agriculture, Belgium. 180pp.

Swennen R. 1990a. Plantain cultivation under West African conditions. A reference manual. International Institute of Tropical Agriculture, Ibadan, Nigeria. 24pp.

Swennen R. 1990b. Limits of morphotaxonomy: names and synonyms of plantains in Africa and elsewhere. Pp. 172-210 in Workshop on the Identification of Genetic Diversity in the Genus Musa. Los Baños, Philippines, September 5-10, 1988 (R.L. Jarret, ed.). INIBAP, Montpellier, France.

Swennen R. & D. Vuylsteke. 1987. Morphological taxonomy of plantain (Musa cultivars AAB) in West Africa. Pp. 165-171 in Banana and plantain breeding strategies, (G.J. Persley & E.A. De Langhe, eds). ACIAR Proceedings 21, Canberra, Australia.

Swennen R., E. De Langhe, J. Janssen & D. Decoene. 1986. Study of the root development of some Musa cultivars in hydroponics. Fruits 41:515-524.

Tennant D. 1975. A test of a modified line intersect method of estimating root length. Journal of Ecology 63:995-1001.

Van Beem J., M.E. Smith & R.W. Zobel. 1998. Estimating root mass in maize using a portable capacitance meter. Agronomy Journal 90:566-570.

Vuylsteke D. 1989. Shoot-tip culture for the propagation, conservation, and exchange of Musa germplasm. Practical manuals for handling crop germplasm in vitro 2. International Board for Plant Genetic Resources, Rome. 56pp.

Vuylsteke D. 1998. Shoot-tip culture for the propagation, conservation, and distribution of Musa germplasm. International Institute of Tropical Agriculture, Ibadan, Nigeria. 82pp.

Vuylsteke D., & D. Talengera. 1998. Postflask Management of Micropropagated Bananas and Plantains. A manual on how to handle tissue-cultured banana and plantain plants. International Institute of Tropical Agriculture, Ibadan, Nigeria. 15pp.

The problem of banana root deterioration and its impact on production58 Distribution of banana roots in time and space: new tools for an old science C.A. Gauggel et al. 59X. Draye et al.

Distribution of banana roots in time and space: new tools for an old scienceXavier Draye1, François Lecompte2 and Loïc Pagès2

AbstractThe distribution of plant roots is the ultimate manifestation of a series of biological processes that operate throughout the plant’s life, in an integrated way, to establish new roots, to regulate the rate of root extension, to influence the direction of root growth and to contribute to root survival. This, combined with the practical difficulties of its estimation, makes root distribution a fairly complex character of plants. Still, root distribution is of paramount significance. It determines the exchange surface between the plant and the soil, the volume of soil influenced by root activity, the depth of soil exploration, the amount of carbohydrates devoted by the plant to root growth and development, the likelihood of infection by soil microorganisms and many other quantitative aspects of the soil/plant system.

Since one of the first reports on the roots of bananas by Fawcett (1921), we have collected more than a hundred contributions in connection with the banana root system. These include descriptions of banana roots and root system, quantitative studies on root growth and its response to environmental factors, anatomical details on root initiation and branching, reports of root distribution and root function (especially water and nutrient uptake), analyses of the interactions between roots and soil pathogens or mycorrhizal fungi and, more recently, germplasm evaluations for some root traits. Fifteen reports are really informative with respect to the spatial distribution of banana roots. They provide a rather unified qualitative picture of how many and where banana root axes (‘cord roots’) are most likely to be found.

Little is known of the lateral roots and root hairs, which contribute significantly to water and nutrient uptake. Compared with the long-lived root axes, they are clearly transient structures. Anatomical data show that they are initiated from root axes following an acropetal sequence. Therefore, the maintenance of a sufficient pool of lateral roots and root hairs depends on the continued production and growth of root axes. As a result, the volume of soil where uptake of water and nutrients occurs is continuously changing as a function of root growth and production. Unfortunately, these time-related components of root distribution remain poorly addressed.

An important question to be raised with regards to root distribution studies is whether a realistic improvement of root traits, be it for performance or sustainability, can be achieved in the future. Available options are the identification, selection or development of improved germplasm, as well as the development/adoption of appropriate cultivation practices. The existence of a significant genetic polymorphism for a number of root traits and the reported tremendous effects of soil conditions on root growth and development support both options. However, there is an urgent need for a list of desirable root properties to achieve better crop performance. This is probably where data are critically lacking. Nonetheless, with existing information we can simulate root architecture in a dynamic way and begin to evaluate current practices. We present an example of the evaluation of sampling practices using a dynamic simulation of a banana root system.

1 Université catholique de Louvain, Croix du Sud 2-11, 1348 Louvain-la-Neuve, Belgium, e-mail: [email protected]

2 INRA, Unité de Recherches en Ecophysiologie et Horticulture, Site Agroparc, 84914 Avignon cedex 9, France, e-mail: [email protected]


Resumen - Distribución de las raíces de banano en tiempo y espacio: nuevas herra-mientas para una ciencia antiguaLa distribución de las raíces en la planta es la manifestación última de una serie de procesos biológicos que ocurren de manera integrada durante la vida de la planta, para establecer nuevas raíces, regular su índice de extensión, influenciar la dirección de su crecimiento y contribuir a su supervivencia. Lo anterior, combinado con las dificultades prácticas para su estimación, hacen que la distribución de la raíz sea una característica bastante compleja de las plantas. Aún así, la distribución de la raíz es de extrema importancia, pues ella es la superficie de intercambio entre la planta y el suelo y de allí depende el volumen de suelo influenciado por la actividad de la raíz, la profundidad de exploración del suelo, la cantidad de carbohidratos utilizados por la planta para el crecimiento y desarrollo de la raíz, la probabilidad de infecciones por micro organismos del suelo y muchos otros aspectos cuantitativos del sistema suelo-planta.

Desde los primeros informes sobre raíces por Fawcett (1921), hemos recolectado más de cien contribuciones relacionadas con el sistema radical del banano. Entre éstas tenemos descripciones de las raíces de banano y de su sistema radical; estudios cuantitativos sobre el crecimiento de la raíz y su respuesta a factores ambientales; detalles anatómicos sobre la iniciación de la raíz y sus ramificaciones; informes sobre distribución y función de las raíces (especialmente absorción de agua y nutrientes); análisis de las interacciones entre raíces y patógenos del suelo u hongos micorrícicos; y más recientemente, evaluaciones de germoplasma para algunas características de las raíces. Quince de éstas contribuciones ofrecen bastante información sobre la distribución espacial de las raíces de banano y presentan un panorama cualitativo bastante unificado de cuántos y donde es más probable encontrar los ejes (“raíces primarias”) del sistema radical del banano.

Se conoce poco sobre las raíces laterales y pelos absorbentes, los cuales contribuyen significativamente a la absorción de agua y nutrientes. Comparados con los ejes de raíz de larga vida, éstas son estructuras transitorias. La información anatómica muestra que se inician en los ejes radiculares, siguiendo una secuencia acropétalo. Por lo tanto, el mantenimiento de reservas suficientes de raíces laterales y pelos absorbentes depende de la producción continua y del crecimiento de los ejes radiculares. En consecuencia, el volumen de suelo utilizado por la raíz para absorber agua y nutrientes se mantiene en constante cambio, como una función del crecimiento y producción de raíces. Desafortunadamente, el tipo de componentes relacionados con el tiempo de distribución de las raíces a recibido poca atención.

Una pregunta importante que se debe plantear con respecto a los estudios de distribución de raíces, sean éstos para propósitos de rendimiento o sostenibilidad, es si será posible lograr en el futuro un mejoramiento realista de las características de la raíz. Algunas opciones disponibles son la identificación, selección y desarrollo de germoplasma mejorado, así como el desarrollo/adopción de prácticas apropiadas de cultivo. La existencia de un polimorfismo genético significativo en varias características de la raíz y el tremendo efecto reportado de las condiciones del suelo sobre el crecimiento y desarrollo de las raíces apoyan ambas opciones. Sin embargo, existe una urgente necesidad de una lista de propiedades deseables en las raíces para obtener mejores rendimientos de cultivo. Este es probablemente el aspecto en el cual se necesita mayor información. Sin embargo, con la información existente podemos simular la arquitectura del sistema radicular en una forma dinámica y empezar a evaluar las practicas actuales. Presentamos aquí un ejemplo de la evaluación de las practicas de muestreo, usando una simulación dinámica del sistema radical del banano.

IntroductionThere is a fairly general agreement about the typical horizontal and vertical distributions of the banana root system: 90% of the roots extend within one meter from the plant, with about 70% of the total root mass of banana plants being confined in the upper 20 to 40 cm of the soil (Fawcett 1921, Champion and Sioussaram 1970, Lassoudière 1978,


Gousseland 1983, Robinson 1985, Araya et al. 1998). One may therefore ask: why a new report on the distribution of banana roots?

A root system is a population of roots of different types and ages that evolve con-tinuously as a result of root formation, growth and senescence. This means that the overall capacity of the root system to keep pace with the various demands placed on it depends on the ability of the plant to maintain an appropriate root population. In this context, what really matters is to understand how the root system responds to environmental constraints and how root form interacts with root function and plant health. The fact that this meeting has been organized is a clear indication that current management practices based on an average picture of banana root distribution remain unsatisfactory.

There is therefore an urgent need for studies of root system architecture that would account for root distribution in space and in time. Only such studies will provide the unifying framework to co-analyse e.g., the processes of soil exploration and resource acquisition, the dynamics of root activity in the rhizosphere and the space-time dyna-mics of plant-interacting soil organisms.

Much has been accomplished since the first report on Musa root architecture at the beginning of the last century (Hartman et al. 1928) and reviews of this work have been recently published (Price 1995, Draye 2002a). The purpose of this paper is to highlight the dynamics of soil exploration by the banana root system and to illustrate ways to use this information to address root distribution in time and space.

Production, growth, directions and survival of the various banana root typesThe functional root system of Musa is based entirely on adventitious root axes, also known as cord roots, arising from a subterranean rhizome (also called the corm) and producing primary and higher-order laterals.

Root axesThe adventitious roots – hereafter referred to as root axes – are the major coarse deter-minants of root distribution. They are relatively straight and cylindrical, up to 5.2 m long (Fawcett 1921) and about 4 to 10 mm in diameter (Acquarone 1930, Riopel and Steeves 1964, Monnet and Charpentier 1965). New axes tend to have increasingly lar-ger diameters (Blomme 2000, Lecompte et al. 2002).

ProductionPositionAdventitious root formation is restricted to a 3-cm long arc near the growing point at the top of the rhizome (Skutch 1932). New roots are therefore produced higher up as the rhizome grows (Summerville 1939, Kwa 1993) and this leads ultimately to the production of roots above the soil surface (Moreau and Le Bourdellès 1963). This phe-nomenon is especially detrimental in soils of low fertility, in windy areas, and under conditions of high parasite load (Braide and Wilson 1980), and may be exaggerated by


a condition referred to as “high mat”, viz. the tendency of corms of successive ratoon cycles to rise above the soil surface (Stover 1972).

A sucker attached to the parent plant may produce 200 to 300 roots before the emis-sion of the first active, non-lanceolate leaf (Champion and Olivier 1961, Robin and Champion 1962). At flowering of the parent plant, 172 to 203 roots may be visible on the tallest sucker (Moreau et al. 1963) and counts exceeding 400 roots have been recor-ded on the subsistent corm of pruned suckers (Turner 1972). It is likely that an interplay of growth substances co-regulate the development of the root system of the parent plant and its suckers and that the suckers contribute to the nutrition of the parent plant and vice-versa as long as they remain attached. It is thus highly recommended that the roots originating from different corms be distinguished from one another when considering root architecture or root function.

TimingThe production of root axes by young Musa plants depends on the type of planting material. On suckers, 10 to 34 roots emerge between 4 and 15 days following planting (Beugnon and Champion 1966, Gousseland 1983, Lavigne 1987). These first axes (pre-formed roots) are most probably growing in the cortex of the corm when the sucker is uprooted and prepared for planting. Suckers that are dug for planting material during periods of reduced root formation may have low numbers of preformed roots (Beugnon et al. 1966). The formation of new roots appears to be interrupted from about 15 days until 75 to 90 days after planting (Champion et al. 1961, Moreau et al. 1963, Beugnon et al. 1966, Gousseland 1983, Lavigne 1987). When the planting material is a section of corm that contains a lateral bud (bits), one to 10 small fibrous roots are initially pro-duced, which do not extend further than the planting hole. Root formation on the new plant starts 3 to 5 weeks after planting and continues thereafter (Turner 1972). Finally, in the case of in vitro plants, the roots produced in the regeneration medium do not grow further in soil or hydroponics and new root formation starts 4 to 5 weeks after planting, at the end of the acclimation phase (Blomme 2000, Stoffelen 2000).

Leaving aside a single study reporting the existence of a second period of root emission at five months in aeroponics (Lavigne 1987), root formation is steady until flowering (Beugnon et al. 1966, Gousseland 1983). It is however highly variable with age and date (Beugnon et al. 1966), promoted by high planting densities (Mohan and Madhava Rao 1984) and can be arrested during periods of low temperatures (Turner 1972, Robinson 1987). About 130 visible roots were counted at 6 months (Summerville 1939), 296 at 7 months (Moreau et al. 1963), 262 to 490 at 8 to 9 months (Beugnon et al. 1966) and 80 and 638 at flowering (Moreau et al. 1963, Turner 1972, Mohan et al. 1984, Lavigne 1987). The plant loses its capacity to form new roots at the end of the vegetative phase and subsequent emission is confined to the suckers (Lavigne 1987).

The scars left by the roots traversing the cortex of the corm can be used to estimate the total number of roots produced, including those that have died (Gousseland 1983). At flowering, the main plant may have produced 400 to 800 cord roots, against 344 to 378 for the tallest sucker (Champion et al. 1961, Robin et al. 1962, Moreau et al. 1963).

The number of roots of the parent plant evolves as leaf area evolves (Gousseland 1983, Lavigne 1987), suggesting a balance between the root and stomatal hydraulic


conductance. The root number is also correlated with size of the plant such as leaf size, bunch size (both number of hands and fingers) and pseudostem diameter (Champion et al. 1961, Beugnon et al. 1966, Lassoudière 1980). In addition, the fruit weight of the parent plant is correlated with the number of roots of the tallest sucker, itself correlated with size of the sucker, mainly the pseudostem circumference. Many of these correla-tions are dependent on place and date.

GrowthThe elongation rate of root axes ranges between 1.2 and 4.0 cm/day (Riopel et al. 1964, Beugnon et al. 1966, Lassoudière 1978, Robinson and Alberts 1989). It is independent of root length (Lavigne 1987, Lecompte et al. 2003). There appears to be a maximum growth rate closely related to the apical root diameter. The degree to which individual roots reach that maximum value depends on carbon allocation to the root apical meris-tem and impedance of root growth (Lecompte et al. 2003). The root growth rate also follows the diurnal (Lassoudière 1978) and seasonal (Robinson 1987) variations of the leaf emergence rate and is adversely affected by mechanical impedance (Gousseland 1983), low pH (Rufyikiri 2000), elevated water tables (Hartman et al. 1928) suboptimal soil temperatures (Robinson et al. 1989) and oxygen deficiency (Aguilar et al. 1998). Based on plausible estimates of root elongation (1.2 to 4.0 cm/day) and lifespan (5-month), a single axis may thus reach a length of 2.6 to 7.5 m, a range that encompasses most observations of maximum root length in good conditions.

DirectionThe first reports on banana root axes suggested that there are horizontal and verti-cal axes (Fawcett 1921, Hartman et al. 1928). It was later argued that the difference between horizontal and vertical roots is one of age and length, and intermediate oblique roots are also present (Acquarone 1930, Summerville 1939). It was further suggested that the earliest -oldest- roots produced at the base of the young rhizome tend to grow downwards, so that roots with lower diameters may be preferentially observed in the subsoil (Lassoudière 1978), although a few vertical roots are regularly found on mature flowering plants (Blomme 2000).

These views were recently challenged by detailed studies of the initial growth direction of roots emerging from the rhizome and of the directional growth of these roots there-after (Lecompte et al. 2002, Lecompte et al. 2003). On one hand, these studies showed that a high proportion of root axes emerge with a low angle to the horizontal and that this proportion is similar for early and late roots. On the other hand, they indicated that roots tend to curve towards the horizontal plane with age. Ultimately, one should account for the large impact of variations in bulk density and mechanical impedance on the soil profile as well as on the localized prevalence of suboptimal soil conditions (Godefroy 1969, Hahn et al. 1996).

SurvivalRoot lifespan can be inferred from a temporal series of counts of the total number of roots produced by a rhizome, including those that have died (see above) and of the number of living roots. Among 110 roots produced during the first 3 months on the parent plant, 50 had disappeared at 3 months and 102 at 7 months (Moreau et al. 1963), indicating that the lifespan of root axes was from about 2 months to more than


4 months. By comparison, glass rhizotron experiments gave estimates of 4 to 6 months (Robinson 1987) and Lassoudière (1980) found live roots on the plant crop at the har-vest of the first ratoon, that is 10 months after their presumed date of emergence. The level of activity of the latter roots is unknown. Interestingly, they were less affected by nematodes than the roots of the follower.

The root axes present at flowering remain alive during the reproductive period (Turner 1970), with a moderate decrease of about 8% (Blomme 2000); and about half of them may still be present at harvest of the follower (Lassoudière 1980). Because root produc-tion vanishes at flowering, the proportion of healthy roots is lower during the reproduc-tive phase (6 to 17%) than at the end of the vegetative phase (16 to 50%; Champion et al. 1961, Robin et al. 1962, Blomme 2000). Furthermore, root length decreased by 40% during the reproductive period, while root number decreased by only 8% (Valsamma et al. 1987, Blomme 2000) indicating that long roots would be more prone to root decay than short ones. This may be related to their being older and to the fact that they explore larger volumes of soil and have higher chances of encountering new hazards. Therefore the maintenance of a high root survival rate after flowering is critical to maintain an adequate root population.

The overall load of factors impeding a root system can be estimated using the distri-bution of root length (number of roots in successive length classes [xi,xi+1[, where w=xi+1-xi is the width of each class). In 10 week-old vitro-plants growing in nutrient solution, that distribution appears to be constant, which would be expected if both new root formation and elongation were constant and roots had not been exposed to hazards. In such conditions, the average root length of young plants in hydroponics sometimes exceeds 75 cm (Swennen et al. 1986, Stoffelen 2000). Conversely, in a study by Moreau and Le Bourdellès (1963) with mature field-grown plants, the distri-bution closely resembled an exponential model: the number of roots in any length class (w=10 cm) was about 65% of that in the preceding class, with 32 to 42% of the roots in the [0,10[ cm length class (i.e. the class includes 0 but excludes 10) and 1% in the [90,100[ cm class (see also Laville 1964, Beugnon et al. 1966). Accordingly, the aver-age root length of field-grown plants during the vegetative phase is thus usually very short, about 30 cm or less (Beugnon et al. 1966, Blomme 2000), and cannot be indica-tive of the volume of soil explored. In theory, the exponential distribution corresponds to a situation where new roots are produced continuously, grow at a constant rate and die only of hazards, which must occur at a constant rate, independent of root age. In conditions of constant root production and growth, the rate of hazards would thus pro-vide a measure of the total biotic and abiotic pressure impeding the root system. The higher the rate, the lower the proportion of roots growing to the next length class. In the study by Moreau and Le Bourdellès, the proportion of roots growing to the next length class were 64, 65 and 66% in three different soils, suggesting that under the conditions stated above the biotic and abiotic pressures in these soils were rather similar.

In an assessment of root density in the top 50 cm layer (weight of roots per weight of soil) using the core sampling technique, Gousseland (1983) reported a 67% decrease in root densities from 30 to 130 cm from the corm. However, after we adjusted this number to the circumference explored by roots at 30 and 130 cm, the total weight of roots present at 130 cm was 42% higher than that of roots present at 30 cm. In a sepa-


rate experiment, the proportion of root weight was not affected by the distance from the pseudostem (Araya et al. 1998). Such problems emphasize the inability of crude esti-mates of root weight data to address issues of root length and root dynamics in soil.

Lateral rootsHaving a usually determinate growth type and a short lifespan, lateral roots do not generally extend far from their parent root. Therefore, while root axes determine the overall exploration around the plant, laterals determine the intensity of local soil explo-ration in the vicinity of these root axes. In this respect, they represent such high num-bers and densities in the soil that they comprise most of the length and area of the root system. In addition, lateral root formation appears to be much more responsive to local soil conditions than is the growth of root axes, making lateral roots the key players in root architectural plasticity and adaptation.

ProductionLateral roots in Musa arise in an acropetal sequence, i.e. new laterals appear closer to the tip than do the previous laterals (Riopel 1966). The general topology of a banana root axis and its laterals resembles that of a herringbone pattern repeated at each order of branching, any root producing lateral roots in a herringbone pattern. However, root branching is affected by rhizosphere conditions as well as by the plant status that may either locally promote or repress the production of lateral roots.

The process of lateral root formation involves the initiation of a primordium in the close vicinity of the parent root apex, followed by a so-called “maturation” phase at the end of which the lateral protrudes from the parent root. The duration of the maturation phase is constant and similar for all root orders of a given genotype, a feature that can be used to estimate the root growth rate of field roots (Lecompte et al. 2001). It was estimated at 3.6 days for ‘Grande naine’ in an andosol in Guadeloupe and at 3.8 days in hydroponics. Lateral roots were consistently found to protrude closer to the tip on roots whose apices are less active (Lassoudière 1978).

First-order laterals usually emerge 10 to 30 cm behind the tip of the root axis (Laville 1964, Riopel et al. 1964, Lassoudière 1978, Blomme 2000). They are 0.5 to 4.0 mm in diameter and reach a density of up to 8 to 10 per cm of root axis in hydroponics (Stoffelen 2000) and in the field (Blomme 2000). The density and length of lateral roots may be slightly depressed in the proximal 10 cm of root axes, probably under the influence and proximity of the corm. Secondary laterals, 1 mm or less in diameter, are often produced on primary laterals (Acquarone 1930, Laville 1964, Blomme 2000) and reach densities of up to 8.7 per cm of primary lateral (Blomme 2000). In nutrient solution, 95% of primary laterals are covered with secondaries (Stoffelen 2000), with a density of 3.0 per cm. Tertiary laterals are rare and less than 200 µm in diameter (Acquarone 1930, Riopel et al. 1964). The pattern of lateral root formation comprises two elements: the restriction to protoxylem-based ranks, and a rather regular spacing within ranks (Charlton 1982), with the logical consequence that the number of lateral roots per unit length of root axis should be proportional to the root diameter (Draye 2002b).

Swennen and colleagues distinguished feeder and pioneer axes, with respectively high and low densities of primary laterals (Winderickx 1985, Swennen et al. 1986).


Compared with the large conical pioneers, feeder axes are cylindrical in shape and have a diameter similar to that of large primary lateral roots (Acquarone 1930). Feeder roots, however, were only observed on suckers (or in vitro plantlets) undergoing a transfer to hydroponics and could not be seen on the suckers produced by plants already grown in hydroponics. Further attempts to distinguish feeders and pioneer root types in the field were also difficult (Blomme 2000). A few observations suggest that they might be replacement roots arising from portions of root axes situated in the cortex of the corm, as a response to a rapid transfer to water medium.

The root tips of the root axes often die when they encounter hard obstacles at right angles. In this case, several lateral roots, almost as large as the parent root, develop behind the dead tip (Acquarone 1930). These laterals have an indeterminate growth (Riopel 1960) and the same internal organzation as primary roots (Acquarone 1930, Riopel 1960, Laville 1964). Because they arise from a region where lateral root primordia are normally present and since their density (number of roots per cm of axis) is similar to that of lateral roots, it is likely that they develop from young primordia of the normal acropetal sequence of lateral root formation. This view is supported by the fact that replacement roots are only produced if the root is damaged within about 2 cm from its apical meristem. Root axes experiencing local mechanical constraints also produce large lateral roots with an apparent indeterminate growth. Typically, where root axes abruptly change direction, large laterals protrude from the external side of the bend. This reaction appears to be proportional to the intensity of the constraint (Lassoudière 1971).

GrowthInformation on the root growth rate of lateral roots remains scanty. First-order lateral roots extend at a rate of 0.7 to 1.5 cm/day (Lassoudière 1978) to reach a final length of 3 to 15 cm (Acquarone 1930, Laville 1964), while secondary laterals can be 0.3 to 4 cm long (Acquarone 1930, Laville 1964, Blomme 2000). It would be interesting to verify whether the relationship between root apical diameter and potential elongation rate, which was found for root axes (see above), also holds for laterals.

The length to which lateral roots grow is probably determined early in their develop-ment. Swennen and colleagues (1986) observed a succession, along water-grown root axes, of segments covered with, alternatively, shorter and longer primary laterals. Given cyclical variations of pH and nutrients which were adjusted weekly, it seems reasonable to propose that the primary laterals have a determinate growth and that their maximum length is determined by rhizosphere conditions prevailing near the tip of the axis at the time the lateral root primordia are initiated (Hackett 1968), as reported with other spe-cies for the response to localized nutrient supply. The production of replacement roots (see above) might be a special case of such a mechanism if one considers not only the influence of rhizosphere conditions but also that of internal conditions.

DirectionAs far as we know, there is no indication of a gravitropic growth for lateral roots. Lateral roots emerge from their parent root at a given azimuth and insertion angle and will adjust their direction of growth in response to soil mechanical anisotropy. Insertion angles would appear to be around 75°, while azimuth angles are determined by the


radial location of the lateral root relative to the protoxylem poles. Since the number of poles is linearly correlated with root diameter, the set of azimuth angles for the laterals of a given root will be proportional to that root diameter.

SurvivalLife spans as short as 6 to 10 days (Lassoudière 1978) to as long as 8 weeks (Robinson 1987) have been reported for first-order lateral roots and are greatly affected by biotic and abiotic constraints. Second-order laterals remain alive for up to 5 weeks (Robinson 1987). Roots of any order, therefore, tend to have a shorter lifespan than their parents. As there is no indication of adventitious lateral root formation in banana, the acropetal pattern of lateral root formation along the parent root implies a parallel acropetal sequence of lateral root decay. A root is therefore comprised of 3 zones: a distal unbranched zone where lateral roots mature, an intermediate branched zone between the regions of lateral root emergence and lateral root decay, and a proximal unbranched zone where lateral roots have died.

The length of the branched zone is a function of the lifespan of laterals and of the growth rate of the parent root. In conditions of continuous root growth, that distance is expected to be constant. However, if the root growth slows, the length of the branched zone will gradually reduce. Ultimately, if root growth stops, the branched zone will gra-dually vanish. This means that the maintenance of a sufficient amount of active lateral roots to achieve efficient resource acquisition depends on a sustained root axis growth. Our inability to reduce conditions that interfere with the growth of root axes (see above) is likely to be a significant component of many soil-related issues of intensive cropping systems. The higher the biotic and abiotic load reducing fine root survival, the more critical the need for conditions that favour the growth of the adventitious roots.

Moving back to root distribution: root architecture modelsAs shown in the preceding section, there exists a significant amount of information on the various processes that determine the architecture of the banana root system. On the whole, this information seems consistent with the various reports of banana root dis-tribution, viz. that 90 % of the roots extend within one meter from the plant, and about 70% of the total root mass of banana plants is confined to the upper 20 to 40 cm of the soil. Still, one would expect process-related information to allow a better understanding of the response of root distribution to the various biotic and abiotic factors than is pos-sible with mere one-shot descriptions of root distribution in specific conditions.

During the last decade, a number of models of root architecture have been developed to facilitate the integration of process-related information into the context of a whole root system in time and space (Pagès et al. 2000). These models use mathematical for-mulae of the process to simulate the activity of every root meristem in time and space, taking account of root production, growth, directional growth and senescence. The coordinates, size, age and topological position of every root segment and meristem are stored and can be analyzed and presented visually at every time step of the simulation. Depending on the deterministic or stochastic formulation of the processes, some degree of variability can be introduced in the simulated root system. Further, the simultaneous consideration of several processes during the simulation makes it possible to connect


different processes, such as growth and branching, or growth of different roots. In this respect, it is interesting to note that whole plant models were recently proposed (Drouet and Pagès 2003).

A prototype model for Musa is currently under development, initially written in the computer language C, and recently ported to object-oriented Java code. It can be pre-sented as a UML (Unified Modelling Language) diagram (Figure 1). The model allows the simulation of a banana field comprising a number of plants. Each plant is initiated as a propagule (e.g. sucker or in vitro plant) that initialises a root and a shoot system. The implementation of the shoot system is limited to a carbon production function taking into account plant age and daily temperature. The implementation of the root system, however, is fairly comprehensive and includes all of the above processes. The code is designed to make it easy to test different implementations of the various proces-

Simulation

Atmosphere

Plot

Plant

Propagule

Shoot System Root system

Root

MeristemSegment

Soil

1

1..n

11 1

1

1..n

1

1

1

1

1

1 1

1

1

1

1

111

1..n0..1

1

Figure 1. Object-oriented UML representation of the Musa root architecture model. Lines connecting different compo-nents (or objects) indicate associations. Diamonds indicate composition relationships. For example: a root system is comprised of 1..n roots, every of which has a unique meristem, 1..n laterals and 1..n root segments that succeed each other. The meristem is linked to the soil because root growth calculation uses soil information (temperature, porosity, mechanical anisotropy).


ses. The current model lacks the ability to simulate sucker development, which would require a clear understanding of allocation rules within the banana stool.

It would be premature to make any statement regarding the precision and relevance of the simulated root systems without intensive testing and further development of the Musa model. For this reason, the present report is limited to a few visual illustrations of model output (Figure 2). However, we are confident that a validated root architecture model of bananas will lead to a better understanding of the root system of bananas in the future.

Figure 2. Illustration of the output of the Musa root model. Vertical (left) and horizontal (right) projections of two simulated root systems at 10, 20, 30, 40 and 50 days (from top to bottom). Grid size = 10 cm x 10 cm.


New methodologies to evaluate root observation methodsThe inherent difficulties in observing and quantifying plant root systems have been a major incentive for the development of a series of root observation methods (Smit et al. 2000). Beyond the complete root system excavation, four methods have been used in bananas: core sampling (Blomme 2000), sector sampling (Araya and Blanco 2001), trench profiles (Delvaux and Guyot 1989) and rhizotrons (Lassoudière 1978). Each of these has its own benefits and drawbacks and makes it possible to estimate a limited number of factors from the plant and/or the root environment. Except for complete root system excavation, the methods involve a sampling strategy that is likely to influence the precision of the estimates and may introduce additional bias. Root architecture models provide a way to estimate the precision and bias with limited input, as well as to identify the conditions under which a method is likely to be robust or error prone.

We present here a fictitious example illustrating how root architecture models can be used to evaluate the precision of root length and root number estimates obtained with the core sampling strategy. The results should not be taken as recommendations for core sampling, although they highlight the importance of some factors affecting the precision of the estimates.

Using the Musa root model, 90 root systems with the same total root length and root numbers have been simulated. These root systems differ only in the spatial arrange-ment of the roots so that the resulting data set would seem appropriate for testing the sensitivity of the core sampling procedure to the major source of variability between plants and around a plant. A virtual sampling was performed for each of the root sys-tems by calculating every 10 days the number and length of root axes and lateral roots in a number of core configurations (number of cores sampled per plant, diameter of the core, distance of the core from the base of the plant). This allowed the computation of average root length and numbers in each sample as well as their confidence limits. These estimates were plotted (Figures 3 to 5) against the true root length and number, and separately for root axes and first order laterals.

This simulation study highlighted the critical effect of the distance between the core and the plant base. At large distances, the length / number of roots in the core are very weak predictors of the total root length / number, which seems consistent with the usual observation that most of the roots are close to the plant. Based on these simulations, a distance of 30 cm or less would be acceptable, yet it is likely that the maximum distance would have been higher had the simulations been run for longer periods. The effect of the core diameter was predictable. At a given distance d from the plant base, large cores intercept a higher proportion of the roots (longer than d) than small cores, and this proportion is directly proportional to the length of the arc intercepted by the core, which normally approximates the core diameter.

Finally, increasing the number of cores may significantly reduce the confidence interval of the estimates (by about 2/3 in these simulations). This example used 25 cm diameter cores sampled at 40 cm from the plant, so that the maximum number of cores was 8. Therefore, confidence limits estimated with n = 8 are the best results that could be achieved with core sampling.


30 cm

15 cm

Adventitious root axes

First order laterals

Total number of root axes Total length of root axes

Total number of lateral roots Total length of lateral roots

Num

berof

root

axes

inprofile

Num

berof

lateralr

oots

inprofile

Leng

thof

root

axes

inprofile

Leng

thof

lateralr

oots

inprofile

10 cm

20 cm





Num

berof

root

axes

inprofile

Num

berof

lateralr

oots

inprofile

Leng

thof

root

axes

inprofile

Leng

thof

lateralr

oots

inprofile

Figure 4. Fictitious core sampling experiment. Effect of the core diameter (from 15 cm to 30 cm by 5 cm intervals). Samples were taken at 20 cm from the plant base.

Figure 3. Fictitious core sampling experiment. Effect of the distance of the core from the plant base (from 10 cm to 90 cm by 10 cm intervals). The core diameter was 25 cm.


The calibration of root architecture models is indeed a prerequisite for using them for recommendations. This calibration should, in principle, be performed repeatedly for every genotype and environmental condition. However, if such calibration data were obtained for a expanding set of genotypes / conditions and if they were adequately documented, they could be used to produce more generic models integrating the effect of genotype and conditions. It is therefore hoped that the quantification of banana root systems in the future will yield new estimates of the dynamic parameters underlying root distribution and will provide good descriptions of the plant and environmental conditions.

PerspectivesBananas and plantains lend themselves to a dynamic modelling of their root architec-ture, a promising step towards an integrated and rational understanding of crop growth and performance. The experience accumulated through more than a century of research has provided a framework to initiate the formulation of a developmental model of Musa root architecture, comprising root formation, branching, elongation and senes-cence, each expressed as a function of genotype, plant status, root age and localized rhizosphere conditions. In principle, this model could be further extended to include as input soil heterogeneity in time and space, genetic correlations between the various root parameters, or also to combine its output with root function data.

n = 1

n = 8





Num

berof

root

axes

inprofile

Num

berof

lateralroo

tsin

profile

Leng

thof

root

axes

inprofile

Leng

thof

lateralroo

tsin

profile

Figure 5. Fictitious core sampling experiment. Effect of the number of cores (n = 1 to 8). The core diameter was 25 cm and cores were sampled at 40 cm from the plant. Continuous line: average values; dotted lines: 95% confidence limits for each number of cores.


While observing architectural differences among genotypes or treatments, one should always attempt to connect these differences to root function. In particular, nutrient and water uptake are not necessarily proportional to root length or surface. In several ins-tances, there even appears to be an excess of roots, the significance of which remains unexplored. In addition, root construction costs have been considered to be proportional to root weight, however it is likely that the amount of photosynthetic energy required to form a unit of root length increases in adverse conditions, as shown with mechanical impedance (Whalley et al. 2000).

With regard to crop performance, the guarantee of optimum root behaviour depends on our ability to take account of root plasticity. The most significant causes of reduced performance in roots in Musa have been identified and the nature of their architectural consequences is being clarified, although the effects of these factors on the late develop-ment of lateral roots remain unknown. It may be wise to establish standard root system descriptors accounting for root plasticity and to be evaluated in defined and rigorously controlled conditions. Given the huge variability of root development under field con-ditions, this may be the only way to bring together the outcome of different research projects. However, before such descriptors can be established, quantitative data on soil heterogeneity (in time and space) prevailing in banana cropping regions are needed.

A root system is a population of roots of different types and ages that evolve dynami-cally according to root formation and root senescence. The overall capacity of the root system to keep pace with the various demands placed on it depends on the ability of the plant to maintain an appropriate root population. In this respect, the importance of appropriate cultivation practices should not be underestimated (Delvaux et al. 1989). Yield decline in plantain, for example, does not occur in backyard or compound gar-dens (Braide et al. 1980). The genetic improvement of the root system remains another possibility. However, in addition to the inherent difficulties of Musa breeding, most questions concerning the characteristics that should be targeted remain unanswered.

AcknowledgementsWe thank the Fonds National Belge de la Recherche Scientifique for their support to XD (Research Associate).

ReferencesAcquarone P. 1930. The roots of Musa sapientum L. United Fruit Company Research Bulletin 26: 831-868.

Aguilar E.A., D.W. Turner, D.J. Gibbs, K. Sivasithamparam & W. Amstrong. 1998. Response of banana (Musa sp.) roots to oxygen deficiency and its implication for Fusarium wilt. Acta Horticulturae 490:223-228.

Araya M. & F. Blanco. 2001. Changes in the stratification and spatial distribution of the banana (Musa AAA cv. Grand Naine) root system of poor, regular, and good developed plants. Journal of Plant Nutrition 24:1679-1693.

Araya M., A. Vargas & A. Cheves. 1998. Changes in distribution of roots of banana (Musa AAA cv. Valery) with plant height, distance from the pseudostem, and soil depth. Journal of Horticultural Science and Biotechnology 73:437-440.


Blomme G. 2000. The interdependence of root and shoot development in banana (Musa spp.) under field conditions and the influence of different biophysical factors on this relationship. PhD Thesis. Katholieke Universiteit Leuven, Leuven, Belgium.

Braide J. & G. Wilson. 1980. Plantain decline: a look at possible causes. Paradisiaca 4:3-7.


Champion J. & P. Olivier. 1961. Etudes préliminaires sur les racines de bananier. Fruits 16:371-374.

Champion J. & D. Sioussaram. 1970. L’enracinement du bananier dans les conditions de la station de Neufchâteau (Guadeloupe). Fruits 25:847-859.

Charlton W.A. 1982. Distribution of lateral root primordia in root tips of Musa acuminata Colla. Annals of Botany 49:509-520.

Delvaux B. & Ph. Guyot. 1989. Caractérisation de l’enracinement du bananier au champ. Incidences sur les relations sol-plante dans les bananeraies intensives de la Martinique. Fruits 44:633-647.

Draye X. 2002a. Banana roots: architecture and genetics. Pp. 261-277 in Plant Roots: The Hidden Half - 3rd ed. (Y. Waisel, A. Eshel & U. Kafkafi, eds). Marcel Dekker, New York.

Draye X. 2002b. Influence of root growth and vascular structure on the distribution of lateral roots. Plant Cell and Environment 25:1463-1474.

Drouet J.L. & L. Pagès. 2003. GRAAL: a model of GRowth, Architecture and carbon ALlocation during the vegetative phase of the whole maize plant - Model description and parameterisation. Ecological Modelling 165:147-173.

Fawcett W. 1921. The Banana. Duckworth & Co., London.

Godefroy J. 1969. Le développement des racines du bananier dans divers sols: relation avec la fertilité. Fruits 24:101-104.

Gousseland J. 1983. Etude de l’enracinement et de l’émission racinaire du bananier ‘Giant Cavendish’ dans les andosols de la Guadeloupe. Fruits 38:611-623.

Hackett C. 1968. A study of the root system of barley I. Effects of nutrition on two varieties. New Phytologist 67:287-299.

Hahn M.L., J.L. Sarah, M. Boisseau, N.J. Vines, D.J. Wright & P.R. Burrows. 1996. Reproductive fitness and patho-genicity of selected Radopholus populations on two banana cultivars. Plant Pathology 45:223-231.

Hartman A.N., A.R. Kieffer, G.F. Gorman & F.W. Harvey. 1928. Banana root studies. A preliminary report on investiga-tions made at Tela, Honduras. United Fruit Company Research Bulletin I:803-816.

Kwa M. 1993. Architecture, morphogenèse et anatomie de quelques cultivars de bananiers. Ph.D. Thesis. Université de Montpellier II, Montpellier, France.

Lassoudière A. 1971. La croissance des racines du bananier. Fruits 26:501-512.

Lassoudière A. 1978. Quelques aspects de la croissance et du développement du bananier ‘Poyo’ en Côte d’Ivoire. Le système radical. Fruits 33:314-338.

Lassoudière A. 1980. Matière végétale élaborée par le bananier Poyo depuis la plantation jusqu’à la récolte du deuxième cycle. Fruits 35:405-446.

Lavigne C. 1987. Contribution à l’étude du système racinaire du bananier. Mise au point de rhizotrons et premiers résultats. Fruits 42:265-271.

Laville E. 1964. Etude de la mycoflore des racines du bananier ‘Poyo’. Fruits 19:435-449.

Lecompte F., H. Ozier Lafontaine & L. Pagès. 2001. The relationships between static and dynamic variables in the description of root growth. Consequences for field interpretation of rooting variability. Plant and Soil 236:19-31.

Lecompte F., H. Ozier-Lafontaine & L. Pagès. 2003. An analysis of growth rates and directions of growth of primary roots of field-grown banana trees in an andosol at three levels of soil compaction. Agronomie 23:209-218.

Lecompte F., A. Vaucelle, L. Pagès & H. Ozier-Lafontaine. 2002. Number, position, diameter and initial direction of growth of primary roots in Musa. Annals of Botany 90:43-51.

Mohan N.K. & V.N. Madhava Rao. 1984. The effect of plant density on the banana root system. South Indian Horticulture 32:254-257.

Monnet J. & J.M. Charpentier. 1965. Le diamètre des racines adventives primaires des bananiers en fonction de leur degré de polyploïdie. Fruits 20:171-173.

Moreau B. & J. Le Bourdellès. 1963. Etude du système racinaire du bananier ‘Gros Michel’ en Equateur. Fruits 18: 71-74.

Pagès L., S. Asseng, S. Pellerin & A. Diggle. 2000. Modelling root system growth and architecture. Pp. 113-146 in Root Methods: a Handbook (A.L. Smit, A.G. Bengough, C. Engels, M. van Noordwijk, S. Pellerin & S.C. van de Geijn, eds). Springer-Verlag, Berlin.

The problem of banana root deterioration and its impact on production74 Distribution of banana roots in time and space: new tools for an old science C.A. Gauggel et al. 75S. Belacázar C. et al.

Price N.S. 1995. Banana morphology - Part I: roots and rhizomes. Pp. 179-189 in Bananas and Plantains (S. Gowen, ed.). Chapman & Hall, London.

Riopel J.L. 1960. Studies on development and wound responses of the roots of Musa ‘Gros Michel’ in relation to Panama disease. PhD Thesis. Harvard University, Massachusetts, USA.

Riopel J.L. 1966. The distribution of lateral roots in Musa acuminata ‘Gros Michel’. American Journal of Botany 53:403-407.

Riopel J.L. & T.A. Steeves. 1964. Studies on the roots of Musa acuminata ‘Gros Michel’ 1. The anatomy and develop-ment of main roots. Annals of Botany 28: 475-494.

Robin J. & J. Champion. 1962. Etude des émissions des racines de la variété du bananier Poyo. Fruits 17:93-94.

Robinson J.C. 1985. Root depth in bananas. Citrus Subtropical Fruit Research Institute Information Bulletin 155:6-8.

Robinson J.C. 1987. Root growth characteristics in banana. Citrus Subtropical Fruit Research Institute Information Bulletin 183:7-9.

Robinson J.C. & A.J. Alberts. 1989. Seasonal variations in the crop water-use coefficient of banana (cv. ‘Williams’) in the subtropics. Scientia Horticulturae 40:215-225.

Rufyikiri G. 2000. Contraintes nutritionnelles chez le bananier (Musa spp.) cultivé en milieux riches en aluminium soluble et conséquences sur sa croissance. PhD Thesis. Université catholique de Louvain, Louvain-la-Neuve, Belgium.

Skutch A. 1932. Anatomy of the axis of the banana. Botanical Gazette 93:233-258.

Smit A.L., A.G. Bengough, C. Engels, M. van Noordwijk, S. Pellerin & S.C. van de Geijn. 2000. Root Methods: A Handbook (1st edition). Springer-Verlag, Berlin.

Stoffelen R. 2000. Early screening of Eumusa and Australimusa bananas against root-lesion and root-knot nematodes. PhD Thesis. Katholieke Universiteit Leuven, Leuven.

Stover R.H. 1972. Banana, Plantain and Abaca Diseases. Commonwealth Mycological Institute, Kew, Surrey, England.

Summerville W.A.T. 1939. Root distribution of the banana. Queensland Agricultural Journal 52:376-392.


Turner D.W. 1970. Banana roots. Agricultural Gazette N.S.W. 81: 472-473.

Turner D.W. 1972. Banana plant growth. 1. Gross morphology. Australian Journal of Experimental Agriculture and Animal Husbandry 12:209-215.

Valsamma M., M. Arvindakshan, P.K. Valsalakumari & N.K. Parameswaran. 1987. Root distribution of banana cultivars under rainfed condition. South Indian Horticulture 35:334-338.

Whalley W.R., J. Lipiec, W. Stepniewski & F. Tardieu. 2000. Control and measurement of the physical environment in root growth experiments. Pp. 75-112 in Root Methods a Handbook (A.L. Smit, A.G. Bengough, C. Engels, M. van Noordwijk, S. Pellerin & S.C. van de Geijn, eds). Springer-Verlag, Berlin.

Winderickx D. 1985. Invloed van pH en temperatuur op het wortelstelsel van banaan en plantaan. PhD Thesis. Katholieke Universiteit Leuven, Leuven, Belgium.

The problem of banana root deterioration and its impact on production74 Distribution of banana roots in time and space: new tools for an old science C.A. Gauggel et al. 75S. Belacázar C. et al.

Development and formation of plantain roots (Musa AAB Simmonds)Sylvio Belalcázar C.1, Franklin E. Rosales2 and Luis E. Pocasangre2

AbstractAll organs that constitute the plantain plant are important and serve a specific function. However, roots and leaves can be considered the two most important organs, because on their physiological behavior and disease status depends the plant’s supply of water and nutrient elements as well as its photosynthetic activity, processes that are closely related to production and yield. A more detailed analysis will probably lead us to consider the root as the most important organ, which unfortunately has not received the attention it deserves by researchers. The plantain root system is spreading and fibrous made up of primary, secondary and tertiary roots, and root hairs. Color is related to age, and may vary from creamy white to yellowish brown. Roots are very fragile early in their life, becoming less so in time, but remaining flexible. The thickness and length of roots are closely related to soil structure and texture. In general, in light soils roots are thin and can reach about 3 m in length, while in heavy soils they are thicker and can reach about 2 m. Roots initiate at the central cylinder or cambium of the corm from which they move to the outer surface through the corm cortex, and emerge either through the nodes or between them, in groups of one, two, three, or up to four roots, without any particular pattern. The root contains the following tissues: epidermis, exodermis, mesodermis, central cylinder or stele, and root-cap. In the vegetative phase, root differentiation is continuous, only interrupted when the lateral bud or follower is separated from the mother plant, to be used as planting material. The follower’s roots continue their growth and development once planted. In this new production unit, root differentiation begins when a second or new corm is formed. The spatial distribution of the root system is a radial horizontal type, where the greatest number of roots is found in the upper 20 cm of soil. The growth and development of the root system is oriented to the soil’s most fertile areas, mostly toward those sites showing high concentrations of organic matter.

Resumen - Desarrollo y formación de las raíces de plátano (Musa AAB Simmonds)Todos los órganos que conforman la planta de plátano son importantes y cumplen con una función determinada. Sin embargo, las raíces y las hojas podrían ser considerados como los órganos más importantes, porque de su comportamiento fisiológico y de su estado fitosanitario, dependen la absorción de agua y elementos nutritivos y la actividad fotosintética, procesos estrechamente relacionados con la producción y el rendimiento. Un análisis más detenido, posiblemente nos lleve a considerar que la raíz sea el órgano más importante de una planta de plátano, la cual infortunadamente no ha recibido la atención que se merece por parte de los investigadores. La morfología radicular del plátano corresponde a un sistema fasciculado y fibroso, conformado por raíces primarias, secundarias, terciarias y los pelos absorbentes. El color guarda relación con su edad y puede variar de blanco cremoso a pardo amarillento. Su consistencia es sumamente frágiles a edad temprana, pero luego con el tiempo se vuelven más resistentes y continúan siendo flexibles. Su grosor y longitud guarda estrecha relación con la estructura y textura del suelo. Generalmente, en suelos livianos son delgadas y pueden alcanzar alrededor de los tres metros de longitud, mientras que en suelos pesados son más gruesas y alcanzan longitudes de aproximadamente dos metros. En cuanto a su anatomía, las raíces se originan en el cilindro central o cambium, desde donde avanzan hacia la superficie externa o corteza del cormo y emergen por lo nudos o bien por los entrenudos, en grupos conformados por una, dos, tres y hasta cuatro raíces, sin ajustarse a ninguna clase de patrón. La raíz está formada por las siguientes

1Honorary Research Fellow, INIBAP-LAC, Armenia, Quindío, Colombia, e-mail: [email protected]; 2 INIBAP-LAC Regional Coordinator and Deputy Regional Coordinator, respectively, CATIE, Turrialba, Costa Rica, e-mail: [email protected]; [email protected]

The problem of banana root deterioration and its impact on production76 Development and formation of plantain roots C.A. Gauggel et al. 77S. Belacázar C. et al.

estructuras: epidermis, exodermis, mesodermis, endodermis, cambium y la cofia. En la fase vegetativa la diferenciación de raíces es un proceso continuo y sólo se interrumpe cuando la yema vegetativa o hijo es separado de la planta madre, para ser utilizada como semilla, en la que las raíces diferenciadas continúan con sus procesos de crecimiento y desarrollo, una vez que se siembran. En esta nueva unidad productiva, el proceso de diferenciación de raíces se inicia tan pronto se forma el segundo cormo. Su distribución espacial puede considerarse como del tipo radial horizontal, encontrándose el mayor número de raíces en la capa de suelo correspondiente a los primeros 20 cm de profundidad. Su crecimiento y desarrollo está orientado hacia las áreas más fértiles del suelo, especialmente hacia aquellos sitios que poseen alto contenido de materia orgánica.

IntroductionThe constant growth of the human population leads to a greater demand for food pro-duction, which for different reasons has not been achieved in several countries, with fatal consequences for people. Public and private institutions, through research centres, continuously seek appropriate economical ways to solve these problems. This leads to the direct and indirect improvement of production factors that can be controlled, such as the development of new cultivars and the establishment and use of new and innovative agronomic practices.

Unfortunately, in the attempt to obtain new, more productive, higher-yielding cultivars, researchers have only considered these factors, forgetting the plant as a whole, such that the organs of the plant that are responsible for its physiological function, have not received the attention and importance they deserve in research. Therefore, when surveys have been conducted, the bunch or perhaps the leaves have been the focus as the most important organ; few agriculturalists have considered the roots as important, reflecting the little attention and interest given to the plant root system.

Among the group of organs that make a plant, the complementary nature and interde-pendent function of roots and leaves should be considered as two of the most important since the plant’s photosynthetic activity, and absorption of water and nutrients, depend on them. The latter functions are closely related to production, and to yield. Among the plant’s organs the root is, without doubt, one of the most important, as it serves not only to anchor the plant to the ground, but also has the primary function of absor-bing and extracting water from the soil, along with essential nutrients necessary for growth, development and plant production. Everything related to growth and produc-tion depends directly on the stage of development, as well as on the cationic exchange capacity of the root system. It also bears mentioning that of the 16 nutrient elements listed as essential for plant growth, 13 are extracted from the soil by the root system.

Morphological description The roots arise from the corm in bundles, they are fibrous and their components have a predetermined growth pattern, as happens with other cultivated species. The roots emerge from the corm either adjacent to, or between, the nodes and either individually or in groups of up to four elements (Belalcázar 1991). According to Cardeñosa (1954), the root system consists of the main roots, the secondary and tertiary roots, and absor-bent hairs. The secondary roots branch off the main roots that are more or less consis-tently thick along their length and do not exceed one cm in diameter. The tertiary roots branch off the secondary roots. The main, secondary and tertiary roots are subsequently


smaller in length and thickness and branch off at right angles. Root hairs grow at the ends of the main, secondary and tertiary roots. The main function of the root hairs, which only grow to a few mm in length, is to absorb water and nutrients. Their capacity to function is directly related to their range.

Root colour, which depends on the age and stage of development, may vary from creamy white to yellowish brown, becoming darker as the root ages. Young roots are fragile, particularly at the tip, which is protected by the root-cap. Roots become more rigid with age, but maintain their flexibility. Root length is influenced by soil texture and structure; the longest roots grow in light sandy loams, and the shortest are found in heavy clay loam soils. Roots in light soils may reach and exceed 3.0 m, while those in heavy soils may barely reach 2.0 m. The converse is true for diameter. At the junction of the rhizome or corm, the diameter varies from 0.4 to 1.0 cm in light soils, while in heavy soils it can be 0.6 to 1.3 cm. Thickness is not constant along the root’s length, but decreases gradually, in both types of soils, from the base to the tip.

The functional or active life of the root is related to different factors, such as the culti-var, presence of insect pests, nematodes and diseases, climate, and chemical and physi-cal properties of the soil. Observations have shown that the functional life of main roots may vary from 5 to 8 months, but apparently functional roots have still been recorded at the time of bunch harvest, at 18 months after planting, and about 11 months after the root differentiation process has finished.

Anatomical descriptionRoots originate at the surface of the central cylinder or cambium, generally in groups of four. Where they originate and develop, the cambium becomes very active, dividing repeatedly at all levels, until forming a primary, cone-shaped meristem with a wide base, from which the buds of 4 adventitious roots develop in acropetal succession. These roots always grow together at the same level. This way there is continuity between the corm’s vegetative buds, and the root.

As the roots grow they have to emerge through the thick cortex of the corm. Unlike other plants, they do not damage the tissues through which they grow, because of the enzymes exuded by the root bud and whose function it is to soften the cortex. However, mechanical damage has been observed in some cases at 1 mm from the end of the emerging root tip. However, at 2 mm from the rhizome surface, the root tip breaks the exodermis (Figure 1).

In general terms, the root is made up of the following structures: epidermis, meso-dermis, endodermis, and central cylinder or stele (Figures 2 and 3), which have the following anatomical construction (Cardeñosa 1954).

Epidermis and exodermisThe epidermis is a layer of cells with slightly thickened lignified walls. Their shape is rounded in a transverse section, and rectangular in a longitudinal one. Extensions to the external part are evidenced at fixed points where the conically shaped root ‘hairs’, or absorbent hairs, originate. The exodermis is composed of 2 to 4 cell layers, quite similar in shape to those of the epidermis, but different in that the cells are a little larger. Also, as the root ages the endodermis walls become thicker and suberize.


Mesodermis and endodermis The mesodermis has a variable number of layers of parenchymatous cells, which vary in shape from the external layers to the inner ones. Cells similar to those in the exodermis are found in the first layer, while the cells of the subsequent layers are larger, spherical, and generally partially disappear pro-ducing ‘watery’ or gas filled cavities. In the inner layers, the cells gradually return to a shape similar to that of the cells in the external layers of the meso-dermis. Intercellular spaces and laticifer vessels are found between the cell layers.

The endodermis is composed of a single layer of thin cells, which in a transverse section are rectangular. The wall adjacent to the stele is very thickened and lignified, while the external portion is thinner. In longitudinal sections, the cells are elongated, and some protuberances can be seen on the surface, as well as small orifices running through the cell wall (Figure 4).

Figure 1. Longitudinal section of a primary root of a ‘Maqueño’ plantain, showing its anatomical structure: A. epidermis and exodermis, B. mesodermis, C. endodermis, D. peripheral areas of the stele (central cylinder), E. secondary lateral root (Photo: R. Cardeñosa).

Figure 2. Cross-section of a ‘Maqueño’ plantain root, showing its anato-mical constitution (Photo: R. Cardeñosa).

Figure 3. Longitudinal section of a ‘Maqueño’ plantain root, showing its anato-mical structures: A. epidermis and exoder-mis, B. mesodermis, C. endodermis, D. stele (central cylinder) (Photo: R.Cardeñosa).


Central cylinder or steleGroups of thin-walled cells, with variable diameters in transverse sections, surrounded concentrically by cells with larger diameters can be found at the periphery of the stele of developing roots but not in mature roots. This is apparently due to meristematic tissue that gives rise to vessels and prosenchyma on its inner layers and to secondary roots at certain external points. The tracheids, which increase in size toward the centre, are dis-tributed across the stele. Several cell layers of the prosenchyma with thick, apparently lignified walls exist between the xylem and the phloem.

The root tipThe root tip does not differ essentially from that common to other monocotyledons. It is composed of several layers of small cells, the quiescent centre (calyptrogen), the epi-dermal stem cells, and the root-cap, which originates before the bud of the root initial has merged from the main root (Figure 5).

Figure 4. Cross-section of a ‘Maqueño’ plantain root, showing the endodermis and other anatomical structures of the stele (Photo: R. Cardeñosa).

Figure 5. Longitudinal section of a ‘Maqueño’ plantain root tip, showing: A.- quiescent centre, B.- root-cap, C.- apical meristem (Photo: R. Cardeñosa).

C

AB


Root differentiation and emissionTaking into account that asexual reproduction is the most common method of propaga-ting plantain, its root system is made up of adventitious, fibrous roots arranged in bun-dles and with a dynamic differentiation and emission process (Schumann 1900). If the climate and soil conditions (particularly water supply) are favourable, new root growth and development begin in the planting material immediately after planting. However, root formation ceases in the corm itself because of the damage caused as it is separated from the parent plant (Belalcázar 1991). More than one hundred roots can be differen-tiated in corms before they are separated from the parent plant (Table 1). These roots emerged, grew, and developed over a period of 6 months. After that time, the number of roots decreased, due mostly to insect pests and nematode attack.

A second corm is formed as a result of the plant growth pattern in the first production cycle, at which point leaf differentiation, vegetative bud and root processes also occur (Figure 6, Table 1). The root system is notably dynamic. Two months after seeding, it had produced 23 roots, and after 9 months, the total number of differentiated roots amounted to 267. Taking account of the number of roots produced by the second corm, these results show that this process begins as soon as the corm is planted and ceases when floral differentiation ends, which occurs when the plant has produced about 50% of its leaves (19±1 leaves) (Belalcázar 1991).

Root spatial distributionRoot system growth and distribution may be conditioned or limited by physical or che-mical factors inherent in the soil, such as porosity, effective depth, degree of aeration, and natural fertility. Therefore, any type of physical or chemical barrier can limit root growth and development (Cardeñosa 1954, Belalcázar 1991). According to the lite-rature, root length may be influenced by soil texture and structure -- reaching greater lengths in light sandy loams than in heavy clay loams.

It has been noticed that most root growth and development occurs in the top 20 cm of soil and that the spatial distribution is horizontal and radial (Table 2).

Table 1. Emission of leaves and the number of roots emerging from the planting material (planted corm) and the new corms arising from it (new corms) of the plantain ‘Dominico hartón’ in the first production cycle. Time Number leaves emitted Number of roots emerging from:

(months) Planted corm New corms

1 2.0* 28

2 4.5 33 23

4 14.0 60 64

6 22.5 102 173

8 28.5 80 237

10 34.5 75 267

12** 38.5 37 241*25 plants average, ** Flowering. Source: Belalcázar et al. (1990).


Figure 6. Diagramatic longitudinal section of the underground corm of the ‘Dominico hartón’ (‘False horn’) plantain, showing the second corm formation (Diagram from A. Belalcázar, 1991). a - growing-point or apical meristem, b - mesodermis (endodermis and cambium), c- cortex (epidermis and exodermis), d - bud, e - groups of emerged roots, f – g - suckers, h - sheaths.

Table 2. Spatial distribution of the number of roots, with depth and horizontal distance from the corm, of the plantain cv. ‘Dominico hartón’ in the first production cycle. Source: Belalcázar (1991). Horizontal distance from the corm (cm)

Depth (cm) 0-20 20-40 40-60 60-80 80-100 Proportion (%)

0-20 105 74 63 33 22 64

20-40 27 24 21 19 8 17

40-60 18 13 8 6 6 11

60-80 11 10 7 5 0 7

80-100 2 0 0 0 0 1

Proportion (%) 35 25 21 11 7

First production cycle

Second production

cycle

e2

e3

h

f

b

g

e1

e1c

c

b

g

c

d

The problem of banana root deterioration and its impact on production82 Development and formation of plantain roots C.A. Gauggel et al. 83M. Araya

Almost two thirds of the roots are concentrated in the topmost soil layer closest to the corm, and the number and percentage of roots gradually decreases as the depth increases, such that at a depth of 1 m, only 1.2% of the emerged roots are present (Table 2). However, because this experiment was conducted in sandy loam soil, with an effec-tive depth over 1.50 m, with good aeration and water and nutrient availability, one may conclude that the root system of the ‘Dominico hartón’ clone plantain is shallow, and additionally, is not influenced by factors related to physical or chemical qualities of the soil, or by water availability.

The exploratory activity of the root system, measured by the number of roots, was very marked horizontally: 82% of the developed roots were recorded in the 0-20 cm depth within a 60 cm radius from the corm (Table 2). The highest percentage of roots (64%) was found in the first 20 cm depth of soil. The difference between this layer, and that from 20-40 cm deep, was very great (47%). Only 17% of the roots were found in this layer.

The above results make it possible to conclude that root system activity is higher in the uppermost soil layers, and that among the tropisms that influence root growth, chemi-tropism has the most influence. The root system is guided to the most fertile zones of the soil that usually correspond to the layers closest to the surface where the greatest amounts of organic matter are found. This organic matter is a source of the majority of essential elements required by the plant, such as phosphorus, which favours root diffe-rentiation and growth and plantain development.

ReferencesBelalcázar S., H. Baena & J.A. Valencia. 1990. Caracterización del ciclo vegetativo del plátano. Pp 150 in Generación

de tecnología para el cultivo y producción rentable del plátano en la zona cafetera central de Colombia. Comité Cafeteros Quindío, ICA, IDRC, INIBAP. Armenia, Colombia.

Belalcázar S. 1991. El cultivo del plátano en el trópico. Feriva. Cali, Colombia. 376pp.

Cardeñosa R. 1954. El género Musa en Colombia, plátanos, bananos y afines. Pacífico. Cali, Colombia. 368pp.

Schumann K. 1900. Musaceae. in Das Pflanzenreich (Engler, A., ed.). Heft 1 Bd. 4:45.

The problem of banana root deterioration and its impact on production82 Development and formation of plantain roots C.A. Gauggel et al. 83M. Araya

Stratification and spatial distribution of the banana (Musa AAA, Cavendish subgroup, cvs ‘Valery’ and ‘Grande naine’) root systemM. Araya1

AbstractThe stratification and spatial distribution of the root system of ‘Valery’ and ‘Grande naine’ bananas was examined at different plant heights, vigour and under different weed control systems in commercial banana plantations in Costa Rica. Changes in the proportion (%) of root fresh weight as horizontal distance from the plant increased, were of similar magnitude in ‘Valery’ plants of different heights. This was also true for soil depth. Root proportions decreased as horizontal distance from the plant increased, partly because the volume of the soil blocks were different at 30, 60 and 90 cm from the plant pseudostem. For the three distances from the pseudostem, the percentage of root fresh weight decreased sharply with increased soil depth. Independent of the distance from the stem, all plants showed the highest root weight (40%) in the top 15 cm of soil. More than 65% of the total root fresh weight was found in the upper 30 cm of the soil, and the first 45 and 60 cm of the soil profile contained more than 79 and 88% of the roots at any plant height. When working with non-flowering ‘Grande naine’ plants reaching 180-190 cm height in sites with plants showing poor, regular or good vigour, no difference was found in total root fresh weight, or in the fresh weight of thick (> 5 mm), thin (1 - 5 mm) or fine (< 1 mm) roots. A difference in the proportion of the fresh weights of the root thickness classes was observed. The distribution was as follows for thick, thin, and fine roots, respectively: 49, 47 and 4% in sites with poor vigour; 63, 33, and 4% in sites with regular vigour and 56, 38, and 6% in sites with plants of good vigour. The total weight of fresh roots and percentage of thick roots decreased as horizontal distance increased at any level of plant vigour. The proportion of fine roots in sites with poor and good vigour was higher at 30-60 cm from the plant base than at the other distances. In sites with poor vigour the total root system did not penetrate more than 75 cm deep while in sites with regular or good vigour it reached 120 cm deep. About 95% of the total root fresh weight was detected in the upper 45 cm in the sites with poor vigour (47% thick, 44% thin, and 4% fine roots). The same layer contained 80% (53% thick, 23% thin, and 4% fine roots) of the root fresh weight in sites with regular and 81% (47% thick, 28% thin, and 6% fine roots) in sites with good plant vigour. No difference in the total root fresh weight was detected in ‘Grande naine’ plants when weeds were controlled for more than 5 years by hand chopping or herbicides. The proportion of thick, thin and fine roots was 66, 30 and 4% respectively. Again, the proportion of root fresh weight of every root type decreased as the horizontal distance from the plant base and soil depth increased. For both weed management systems, the total root system was found in the upper 60 cm, with 85% in the first 30 cm layer. Even though no statistical difference was found in bunch weight, the manual weed control plot yielded bunches 1.4 kg heavier than those where chemical weed control was used.

Resumen - Estratificación y distribución espacial del sistema radical del banano (Musa AAA, subgrupo Cavendish, cvs ‘Valery’ y ‘Grande naine’)La estratificación y la distribución espacial de sistema radical de los cvs ‘Valery’ y ‘Grande naine’ fueron examinadas a diferentes alturas de planta, vigor y bajo dos sistemas diferentes de control de malezas en plantaciones comerciales de banano en Costa Rica. Los cambios en los porcentajes del peso fresco de las raíces fueron de similar magnitud en ‘Valery’ conforme se aumentó la distancia horizontal desde la base de la planta, en plantas de diferentes alturas. Los cambios entre los

1 Corporación Bananera Nacional (CORBANA S.A.) Apdo 390 7210 Guápiles, Costa Rica. e-mail: [email protected]

The problem of banana root deterioration and its impact on production84 Stratification and distribution of the banana root system C.A. Gauggel et al. 85M. Araya

porcentajes con las profundidades de suelo también fueron de similar magnitud. El porcentaje de raíces decreció conforme aumentó la distancia horizontal desde la base de la planta, debido en parte a que el volumen de los bloques de suelo fue diferente a 30, 60 y 90 cm. Para las tres distancias horizontales, el porcentaje de raíces decreció con la profundidad del suelo. Independientemente de la distancia al tallo, todas las plantas mostraron el mayor peso (40%) de raíces en los primeros 15-cm de suelo. Más del 65% del total de raíces se encontró en los primeros 30 cm de suelo y en los primeros 45 y 60 cm del perfil del suelo se encontró más del 79 y 88% de las raíces, en cualquiera de las alturas de planta evaluadas. Cuando se trabajó con plantas no florecidas de ‘Grande naine’ de 180-190 cm de altura en áreas con pobre, regular y buen desarrollo, no se encontraron diferencias en el contenido fresco del total de raíces ni en el peso fresco de las raíces gruesas (> 5 mm-d), delgadas (1 - 5 mm-d) o finas (< 1 mm-d). Se observó diferencia en la proporción del peso fresco de los diferentes grosores de raíces. La distribución fue la siguiente para raíces gruesas, delgadas y finas, respectivamente: 49, 47 y 4% en áreas con plantas de pobre desarrollo; 63, 33 y 4% en sitios con plantas de regular desarrollo y 56, 38 y 6% en sitios con plantas de buen desarrollo. El porcentaje del total de raíces y raíces gruesas decreció conforme se incrementó la distancia horizontal en cualquier desarrollo de planta. La proporción de raíces finas en sitios con plantas de pobre y buen desarrollo fue mayor a 30-60 cm de la base de la planta que a las otras distancias. En los sitios con plantas de pobre desarrollo el total del sistema radical se encontró sobre los 75-cm de profundidad, mientras en sitios con plantas de regular y buen desarrollo el sistema radical alcanzó los 120-cm de profundidad. Cerca del 95% del peso fresco del total de raíces se detectó en los primero 45 cm del suelo en los sitios con plantas de pobre desarrollo (47% de gruesas, 44% de delgadas y 4% de finas). En esa misma capa se encontró un 80% (53% de gruesas, 23% de delgadas y un 4% de finas) en el sitio con planas regulares y un 81% (47% de gruesas, 28% de delgadas y 6% de finas) en el área con plantas de buen desarrollo. No se encontraron diferencias en el peso fresco del total de raíces de ‘Grande naine’ cuando el control de malezas se realizó por más de 5 años con chapeas o herbicida. La proporción de raíces gruesas, finas y muy finas fue de 66, 30 y 4%, respectivamente. Nuevamente el porcentaje del peso fresco de cada tipo de raíz decreció conforme aumentó la distancia horizontal de la base de la planta o aumentó la profundidad del suelo. Para ambos sistemas de manejo de malezas, el total del sistema radical se encontró en los primeros 60 cm de suelo, con 85% en los primeros 30 cm. Aún cuando no se detectaron diferencias estadísticas en el peso de los racimos, las parcelas con control manual de malezas produjeron racimos con 1.4 kg más de peso.

Introduction As stated by Stover and Simmonds (1987) and Price (1995), descriptive studies and scientific research on the banana (Musa AAA) root system are quite limited, especially research on root physiology, metabolism and interactions between the soil and soil microorganisms. This lack of research is probably related to the complexity of the chemical, physical and biological interactions that occur between roots and the surroun-ding soil environment (McCully 1999). Some work has been undertaken to understand the development (Skutch 1932, Beugnon and Champion 1966, Lavigne 1987, Blomme and Ortiz 1996, Keshava and Iyengar 1997, Draye et al. 1999), function (Mohan and Rao 1984), and distribution (Riopel 1966, Sioussaram 1968, Champion and Sioussaram 1970, Avilán et al. 1982, Gousseland 1983, Mathew et al. 1987, Price 1995, Araya et al. 1998) of the banana root system. In an effort to estimate the root system based on measurements of aerial parts, Blomme et al. (2001) found that leaf area, pseudostem circumference and height of the tallest sucker were the best indicators.

In the present study, the stratification and spatial distribution of the root system of ‘Valery’ and ‘Grande naine’ bananas was examined at different plant heights, vigour and under different weed control systems in commercial banana plantations in Costa Rica.


Formation and functions of the root systemThe banana root system follows the same overall arrangement of many other mono-cotyledons (Esau 1965). It consists of adventitious roots (Skutch 1932, Price 1995) that usually arise in groups of three or four from a common primordium within the corm, and emerge from the corm individually or in groups of two, three or four roots (Figure 1). Roots are produced continuously until flowering (Beugnon and Champion 1966, Mohan and Rao 1984, Lavigne 1987). On each first order adventitious root, sometimes called a primary root, root axis or cord root, a system of secondary roots may be formed, and on this, tertiary roots develop (Riopel 1966, Swennen et al. 1986, Sandoval and Müller 1999), each order being progressively thinner and shorter than the former (Figure 1). Roots of the second and third orders originate in the protoxylem near the root tips of the first and second order roots, and continue to be produced as the first and second order root extends through the soil (Stover and Simmonds 1987, Robinson 1996). When the tip of a first order root is damaged by biotic or abiotic fac-tors, two or three second order laterals may develop from behind it into long cord roots (Lassoudière 1978, Kobenan et al. 1997).

Banana roots are responsible for anchorage, water and nutrient supply, and exchange of various growth substances with the shoots, including plant growth hormones such as: cytokinins, abscisic acid, ethylene, gibberellins, auxins and jasmonic acid (Barber 1992, Itai and Birnbaum 1996).

Different types of roots have distinctly different ion uptake patterns, both spatially and temporally, and uptake changes with age and distance from the root tip (Waisel and

Figure 1. Left: root system of Musa AAA cv. ‘Grande naine’ originated from in vitro propagation material showing the first order adventitious root (root axis, cord root or primary roots) that emerged from the corm of a plant excavated and extracted from the field. Right: First order adventitious root (root axis, cord root or primary root) (A) where second (B) and third order (C) roots develop.


Eshel 1992, Zobel 1992, Gao et al. 1998). Uptake depends on the root surface area, the kinetics of nutrient uptake by the root and the rate of nutrient supply to the roots by mass flow or diffusion (Barber 1992). Nutrients move to the root surface radially, which means that the nutrient supply to the root per unit of root surface area is greater for fine roots than for coarse roots. Moreover, young roots are generally considered to have a higher activity of nutrient uptake than old roots (Gao et al. 1998). This causes finer roots to have on average a higher nutrient concentration in solution at the root surface and a greater uptake rate. Atkinson (1989) proposed that increased root system size would be advantageous for soil resource use.

Robinson (1987), working with the Cavendish subgroup banana cv. ‘Williams’, found that cord roots were functional for 4-6 months, whereas secondary and tertiary roots for 8 and 5 weeks, respectively, and root hairs for about 3 weeks, before rotting away.

Effect of abiotic factors on the root systemFunctional and healthy roots are essential for good production. Therefore, most agri-cultural investment (soil preparation, fertilization, irrigation scheduling, pest control) is aimed at providing optimal conditions for root growth. Bananas have been regarded as an adaptable crop within certain, well-defined climates and soil types. The recommen-ded soils for banana are slightly acidic, deep, rich and well drained, such as clay loam or loam soils (Lahav and Turner 1992, Delvaux 1995, Borges et al. 1999). Soil type, compaction, drainage, and variations in climate have a strong influence on the develop-ment of the root system (Avilán et al. 1982, Stover and Simmonds 1987, Barber 1992, Klepper 1992, Robinson 1996, Blomme et al. 2002, Dorel 2002). Great variation exists in the spatial and temporal distribution of resources in the soil, and since resources vary in mobility, rooting patterns may also vary. Heavy, compact, poorly-drained soils and water tables at depths of less than 1 m severely limit root extension, and yields are depressed accordingly (Stover 1972, Lassoudière 1978, Stover and Simmonds 1987, Robinson 1996).

Godefroy (1969) found a deeper rooting pattern for plantain (Musa AAB) cultivated on a sandy clay soils. Araya and Blanco (2001) observed that in soils where plants had poor development, the root system only reached 75 cm, with 99% of their roots in the first 60 cm. This was related to soil compaction beyond 60 cm depth, which is usually associated with less pore space and poor drainage, thus impeding root growth (Bennie 1996). Soil structure alters root distribution (Irizarry et al. 1981, Bennie 1996, Blomme et al. 2002). In bananas, Weert (1974) found that in soil with less than 5% macropores root penetration was drastically reduced. Similar results were reported by Avilan et al. (1982), where soil with 3 and 5% of macropores in the first 30 cm depth showed root deformation and reduced root weight. Blomme et al. (2002) found that ploughing the soil to depths of up to 30 cm improved the root system of 6 genotypes belonging to four Musa spp. groups. In other crops, it is well documented (Wiersum 1957, Bennie 1996, Charlton 1996, Fitter 1996, Goodman and Ennos 1999, Pezeshki et al. 1999) that soil density, moisture, pore structure, soil aeration, soil nutrient content and nutrient supply restrict root growth and elemental uptake.

Soils with poor drainage affect Musa root development and yield, because depth to the watertable limits root penetration (Godefroy 1969, Lassoudière 1971 and 1978, Irizarry et al. 1980, Avilan et al. 1979 and 1982). Aguilar et al. (2003) demonstrated that even


short-term O2 deficiency in the banana root zone could, within minutes, adversely affect root aeration and consequently, within minutes to hours, root functions such as net solute transport and root elongation. Furthermore, the reduced products of the che-mical and microbiological transformation due to the depletion of oxygen may behave as toxins depending on their concentrations (Drew and Stolzy 1996). Differences in the development of aerenchyma in banana roots has been reported by Aguilar et al. (1999), who mentioned that this could be useful in selecting cultivars for flood tolerance. Irizarry et al. (1980) found the best plantain yield when the water table was lowered to 36 cm depth. Also they recorded an increase in roots per volume of soil with a free water table at 48 cm depth. Ghavami (1976) found that bunch weight increased and the root system was deeper as depth to the water table increased to 120 cm. Under soil water deficit, a reduction in fresh and dry weight of banana fruits was found by Hegde and Srinivas (1989). Bhattacharyya and Madhava (1988a, b, 1989, 1990, 1992) and Robinson et al. (1991) found that manipulation of the soil water regime through the use of soil covers, such as banana trash and sugarcane leaf trash increased root cation exchange capacity, enhanced banana leaf production, leaf relative water content, and the C/N ratio in the leaves, and reduced the shooting-harvest interval.

In some conditions, the application of mulch stimulated better root development, which resulted in better anchorage, and improved water and nutrient absorption (Swennen and Wilson 1985). Bhattacharyya and Madhava (1986) found that banana plants, grown with a soil cover of banana trash and sugarcane leaf trash, had shallow root systems, but enhanced yield. One benefit of the use of banana trash and sugarcane leaf trash as a soil cover was the reduction and stabilization of the soil temperature down to 15 cm deep (Bhattacharyya and Madhava 1986, Robinson et al. 1989). It is known that soil temperature affects root growth, and an optimum exists for each crop (McMichael and Burke 1996). The dampening effect on soil temperature may be reproduced with manual weed control in banana plantations. Talwana et al. (2003) hypothesized that lower soil temperatures in mulched areas may slow the feeding activity and repro-duction of R. similis and thereby reduce root damage. Manual weed cutting in banana plantations supplies sufficient ground cover and when done periodically provides good mulch, which is also a source of nutrients and organic matter (De La Cruz et al. 2001). Ruhigwa et al. (1995a) observed that mulch and weeds on the soil surface stimulated ramification of plantain roots. Refuse from harvested banana plants and from deleafing for black Sigatoka (Mycosphaerella fijiensis) control is another source of mulch, and also of nutrients and organic matter, and thus acts as fertilizer (Ruhigwa et al. 1995a, Vargas and Flores 1995). Ruhigwa et al. (1995a, b) found that plantain with Pennisetum purpureum mulch had a shortened fruiting cycle and gave the highest yield. Also, Salau et al. (1992) found that mulching stimulated vegetative growth and increased yield.

The beneficial effects of mulching are attributed in general to the improvement of physical, chemical and biological soil properties. However, the use of mulch provides other benefits such as reduced soil erosion, reduced temperature fluctuation, reduced evaporation rate, reduced nutrient leaching and increased water infiltration (Obiefuna 1991, Salau et al. 1992, De La Cruz et al. 2001). Additionally, decreases in pesticide application might reduce soil, air and human pollution.


Effect of biotic factors on the root systemGenomic group, ploidy level and the health of the plant affect the root system (Swennen et al. 1986, Stover and Simmonds 1987, Blomme 2000, Blomme et al. 2000, 2003). Moreover, the age of the plant and cropping patterns can modify the root distribution of any crop (Mohan and Rao 1984, Barber 1992, Klepper 1992). Also, the extension of the feeding zone varies widely among species and clones (Swennen et al. 1986 and 1988, Mathew et al. 1987, Klepper 1992). In local conditions, when comparing six genotypes and working with excavated plants (Figure 2), ‘Yangambi Km5’ presented the highest

Excavatedarea

30 cm

30 cm

60 cm

60 cm

90 cm

90 cm

120 cm

corm

Figure 2. Left: schematic representation of the excavated area. Transverse section A-A. Right: Evaluation of a plant in the field.

(3584 g/plant) and ‘Valery’ the lowest (892 g/plant) amount of roots (Figure 3). The dis-tribution of thick (> 5 mm), thin (1-5 mm) and fine (> 1 mm) roots varied within genoty-pes and only thick roots varied among genotypes. In any genotype, the highest proportion was of thin roots (61-77%) followed, with the exception of ‘Valery’, by thick and fine roots (Figure 4). The horizontal and vertical root distribution follows the same pattern in all genotypes, decreasing as distance from the plant base or soil depth increased (Figure 5). Blomme et al. (1999) also reported that ‘Valery’ has a small root system.

Vesicular-arbuscular mycorrhizae (VAM) are a type of symbiosis between plant roots and fungi. Most cultivated plants, including bananas, are hosts of mycorrhizae (Girija and Nair 1988, Iyer et al. 1988, Villalobos et al. 2001, Azcón et al. 2002). Mycorrhizae improve transplant recovery and survival, enhance growth rate, and increase drought resistance, effects that are thought to be directly or indirectly due to the ability of the fungus to aid in uptake of phosphorus (John 1988, Blal et al. 1990, Jaizme et al. 2002). One approach is attempting to develop VAM as a nematode control tool in bananas (Umesh et al. 1988, Parvatha et al. 2000, Elsen et al. 2002) that could be included in a strategy for the integrated management of banana nematodes.

Nematodes are important pathogens in bananas and in many countries rank second after black Sigatoka. However, they will not be included here, because interesting


book chapters have already been written on this subject (Gowen and Quénéhervé 1990, Gowen 1995, De Waele 2000, De Waele et al. 2000, Gowen 2000a, b, Sarah 2000). Moreover, there are two presentations about nematodes in these proceedings. Associations between nematodes and other organisms have been noted in bananas (Pinochet and Stover 1980, Aguilar 2002, Batlle and Pérez 2002). In local conditions, Trichoderma spp., Fusarium spp., Penicillium spp., and Cylindrocladium spp. were

0

500

1000

1500

2000

2500

3000

3500

4000

4500

FHIA-18 FHIA-23 GrosMichel

PJB Valery YG-km5

Musa genotypes

Totalfresh

root

weigh

t(g)

Figure 3. Total root fresh weight in the excavated area of banana (Musa spp.) genotypes. Data are mean ± standard error of 10 plants.

100

90

80

70

60

50

40

30

20

10

0FHIA-18 FHIA-23 Gros

MichelPJB Valery YG-km5

Musa genotypes

Percen

tage

Thick (>5 mm-d) rootThin (1-5 mm-d) rootFine (<1 mm-d) root

Figure 4. Distribution of root types by genotype. Data are mean ± standard error of 10 plants.


associated with banana roots showing damage mainly by R. similis (Aguilar 2002). Cylindrocladium spp. has also been associated with R. similis root damage in banana plantations in Martinique and Guadeloupe (Loridat and Ganry 1991). Stover (1966) and Pinochet and Stover (1980), monitoring banana root nematode lesions from plan-tations in Central and South America, Somalia and the Philippines, reported evidence of Fusarium solani, F. moniliforme, Cylindrocarpon musae and Acremonium stroma-licum. In Lebanon, F. oxysporum and Rhizoctonia solani were associated with banana root damage caused by Helicotylenchus multicinctus or Meloidogyne incognita (Sikora

CHD 60-90HD 30-60HD 0-30

CHD 60-90HD 30-60HD 0-30

CHD 60-90HD 30-60HD 0-30

CHD 60-90HD 30-60HD 0-30

CHD 60-90HD 30-60HD 0-30

CHD 60-90HD 30-60HD 0-30

0

5

10

15

20

25 FHIA-18

0

5

10

15

20

25PJB

0

5

10

15

20

25 FHIA-23

0

5

10

15

20

25Valery

0-15

15-30

30-45

45-60

0

5

10

15

20

25

Soil depth (cm)

Gros Michel

0-15

15-30

30-45

45-60

0

5

10

15

20

25

Soil depth (cm)

YG-km5

EB

DA

FC

Percen

tage

withrespec

tto

thetotalroo

tweigh

tin

theex

cavatedarea

Percen

tage

withrespec

tto

thetotalroo

tweigh

tin

theex

cavatedarea

Figure 5. Vertical and horizontal distribution (%) of banana (Musa spp. genotypes) roots. HD=horizontal distance from the plant pseudostem. Data are means of 10 plants.


and Schlösser 1973). However, there are some endophytic fungi that colonize the roots and protect them against biotic or abiotic factors (Pocasangre et al. 2000). These authors found five non-pathogenic strains of Fusarium that were able to reduce R. similis in banana cultivars. More information about this approach is presented in these proceedings.

Spatial root distribution in ‘valery’ plants differing in height Roots generally spread over 2-3 m and up to 5 m from the plant (Riopel 1966, Avilán et al. 1982, Gousseland 1983) and over 1 m in depth (Price 1995). However, most of the root system occurs within a 60 cm radius of the stem and within 30 cm of the soil surface (Sioussaram 1968, Champion and Sioussaram 1970, Avilán et al. 1982).

Araya et al. (1998), working with excavated plants (Figure 2) of Musa AAA cv. ‘Valery’, did not find an effect of plant height on the spatial distribution of the roots, expressed as the proportional distribution of fresh weight. This means that changes in the proportion of root fresh weight, as horizontal distance from the plant increased, were of similar magnitude in vegetative plants of 120 (2 leaves), 150 (4 leaves), 180 (6 leaves) or 210 (8 leaves) cm high or 390 cm high in plants within 1 to 3 days from bunch (inflorescence) emergence. This was also true for soil depth. They found in the upper 30 cm of the soil profile a higher proportion of root fresh weight at 0-30 cm horizontal distance from the pseudostem than further away (Figure 6). The proportion

0-15

15-30

30-45

45-60

60-75

75-90

90-105

105-120

0

2

4

6

8

10

12

14

16

Freshroot

weigh

t(%)

Soil depth (cm)

HD 60-90HD 30-60

HD 0-30

Figure 6. Spatial distribution of Musa AAA (cv ‘Valery’) roots. HD = horizontal distance from the plant pseudostem. Data are mean of 20 plants.


of the roots decreased as horizontal distance from the plant increased, partly because soil block volume was different at 30, 60 and 90 cm from the pseudostem. In the smal-lest plant (120 cm high) the volume of the soil blocks at 60 and 90 cm from the plant base was 2.3 and 3.6 times the volume at 30 cm, while in the plants close to bunch emergence, it was 2.0 and 3.1, respectively. For the three distances from the stem, the proportion of root weight decreased sharply with soil depth (Figure 6). Independent of the distance from the stem, all plants showed the highest root weight in the top 15 cm of the soil (40%). More than 65% of the total root fresh weight was found in the upper 30 cm of soil and more than 79% and 88% of the roots were found in the first 45 and 60 cm of the soil profile, respectively, at any plant height. The total root fresh weight in the excavated area (0-90 cm from the stem and 0-120 cm depth) was 1.2 kg for plants 120 cm high, 1.0 kg for 150 cm, 1.1 kg for 180 cm, 1.5 kg for 210 cm high and 1.2 kg in plants just before bunch emergence.

Stratification and spatial root distribution in ‘Grande naine’ plants differing in vigourAraya and Blanco (2001) did not find differences in total root fresh weight or in the root fresh weight of thick (> 5 mm), thin (1 - 5 mm), or fine (< 1 mm) roots per plant among sites with plants of different vigour and bunch weight (Figure 7). Changes were obser-ved in the proportion of root types with plant development. In sites with good plant vigour there was 56% of the root fresh weight in thick roots, 38% in thin roots and 4% in fine roots. For the sites of moderate vigour, the distribution was 63% for thick roots,

1000

900

800

700

600

500

400

300

200

100

0Total root Thick roots Thin roots Fine roots

Root types

Weigh

t(g)

Poor plants

Regular plants

Good plants

<1 mm-d1 mm-d> 5 mm-d

Figure 7. Root fresh weight by root type in the excavated area of bananas (Musa AAA cv. ‘Grande naine’) with poor, regular and good plant vigour. Data are mean ± standard error of 10 plants.


33% for thin roots and 4% for fine roots. These values changed to 49% thick roots, 47% thin roots and 4% fine roots in sites with poor vigour. The proportion of total root fresh weight, per volume of soil, (Figure 8A) decreased as horizontal distance from the plant increased at any plant development site. When roots were partitioned in classes, the same pattern was observed for thick roots (Figure 8B). However, the proportion of thin roots (Figure 8C) in poor and good plants was larger at 30-60 cm from the plant base than at the other distances. This was also true for fine roots (Figure 8D) in the plants with good development.

In plants with poor vigour, no roots penetrated deeper than 75 cm while in plants with regular and good vigour they reached 120 cm depth (Figure 9A). About 95% of the total root fresh weight was detected in the upper 45 cm of soil in plants of low vigour (47% thick, Figure 9B; 44% thin, Figure 9C; and 4% fine, Figure 9D). In the same layer there was 80% (53% thick, 23% thin, and 4% fine) in sites with plants of regular vigour and 81% (47% thick, 28% thin, and 6% fine) in sites with vigorous plants. The total root fresh weight in the excavated area (0-90 cm from the pseudostem and 0-120 cm depth) was 0.95 kg for the plants of poor vigour, 0.9 kg for the moderate and 0.76 kg for plants with good vigour.

Percen

tage

withrespec

tto

thetotalroo

tweigh

tin

theex

cavatedarea

0-30

30-60

60-90

0

10

20

30

40

50Total root

0

10

20

30

40

50

Fine (<1 mm) roots

0

10

20

30

40

50

Thin (1-5 mm) roots

Poor plantsRegular plantsGood plants




0

10

20

30

40

50

Thick (> 5 mm) roots

A

B

C

D

Horizontal distance (cm)

0-30

30-60

60-90

Horizontal distance (cm)

Figure 8. Horizontal distribution (%) of roots from bananas (Musa AAA cv. ‘Grande naine’) of poor, regular and good plant vigour. Data are means of 10 plants.


Stratification and spatial root distribution in ‘Grande naine’ under two weed management systemsNo difference was found in total root fresh weight between sites with chemical (767 g/plant) or manual weed control (781 g/plant). Also, there was no difference between management systems in any root thickness class, but root types differed between them (Figure 10). The average root fresh weight for both weed management systems was 516 g of thick roots per plant, 222 g of thin roots and 33 g of fine roots per plant. No differences were observed between weed management systems in the horizontal or vertical proportional distribution of fresh thick, thin or fine roots per plant. The proportion was 66% for thick, 30% for thin and 4% for fine roots. The proportion of root fresh weight of every root type decreased as horizontal distance from the plant and soil depth increased (Figure 11A-C). In both weed management systems, the total root system was found within the first 60 cm of soil depth. About 85% of the total root fresh weight was detected in the upper 30 cm, and 44% of the root system was located in a 30 cm radius around the pseudostem.

Percen

tage

withrespec

tto

thetotalroo

tweigh

tin

theex

cavatedarea

Total roots

Fine (<1 mm) rootsThin (1-5 mm) roots


Thick (> 5 mm) roots

Horizontal distance (cm) Horizontal distance (cm)

0-15

15-30

30-45

45-60

60-75

75-90

90-105

105-012

0-15

15-30

30-45

45-60

60-75

75-90

90-105

105-012

0-15

15-30

30-45

45-60

60-75

75-90

90-105

105-012

0-15

15-30

30-45

45-60

60-75

75-90

90-105

105-012

0

10

20

30

40

50

0

10

20

30

40

50A

0

10

20

30

40

50

0

10

20

30

40

50

B

C D

Figure 9. Vertical distribution (%) of roots from bananas (Musa AAA cv. ‘Grande naine’) of poor, regular and good vigour. Data are means of 10 plants.


0-15

15-30

30-45

45-60

HD 0-30HD 30-60HD 60-90

0

5

10

15

20

25

Soil depth (cm)

0-15

15-30

30-45

45-60

Soil depth (cm)

0-15

15-30

30-45

45-60

Soil depth (cm)

Thick (> 5mm-d) roots

HD 0-30HD 30-60HD 60-90

0

5

10

15

20

25

Thin (1 - 5 mm) roots

HD 0-30HD 30-60HD 60-90

0

5

10

15

20

25

Fine (< 1mm-d) roots

C

BA

Percen

tage

withrespec

tto

thetotalroo

tweigh

tin

theex

cavatedarea

Figure 11. Vertical and horizontal distribution (%) of banana (Musa AAA cv ‘Grande naine’) roots from either chemical or manual weed control. HD= horizontal distance from the plant pseudostem. Data are means of 20 plants.

0

100

200

300

400

500

600

700

800

900

Total roots Thick roots Thin roots Fine roots

Root types

Weigh

t(g)

Hand cutting

Chemical control

Figure 10. Fresh root weight by root type in the excavated area of bananas (Musa AAA cv. ‘Grande naine’) with chemi-cal or manual weed-control. Data are mean ± standard error of 10 plants.


Bunches from the plot where weeds were controlled manually were 1.4 kg heavier than those from the plot with herbicide applications. No correlation was found between bunch weight and either total root weight, or root weight by thickness class or the sum of root types per soil block.

Practical implications of the stratification and spatial distribution of the banana root system Commercial banana farms in Costa Rica are usually larger than 80 ha and in many cases it is common to observe plantations of 400 ha or more. The complete area in a given farm is treated in the same way. This means that fertilizer application, nematode, weed and black Sigatoka control, drainage and other cultural practices are equal across the farm. However, differences in plant development and yield are observed, which should give clues to managing those sites accordingly.

Independently of the research focus, the major part of the banana root system occurs in the upper part of the soil. This is in agreement with other results (Sioussaram 1968, Champion and Sioussaram 1970, Avilán et al. 1982, De Moura et al. 1986, Martinez and Ortega 1988, Soto and Ruiz 1992, Price 1995). Due to this superficial root system, mechanical cultivation should be avoided. Generally, in commercial banana farms, weed control is based on herbicide application. This reduces weed control costs com-pared with manual control, but according to Stover and Simmonds (1987) it increases erosion, especially in high-rainfall areas as in our study conditions. Moreover, the post-emergence herbicides should be applied with caution to prevent the product from ente-ring the soil and killing the roots. Since herbicides have an impact on the environment and because the demand for pesticide-free products is increasing, manual weed control is becoming an alternative.

No difference in root weight between chemically and manually controlled plots indi-cated that weeds used as cover crops did not compete for space with the banana root system nor did they affect root stratification. Bunches from the manual weed control plot were 1.4 kg heavier compared with chemically-treated plots. This means that manual weed control did not affect yield, which is in partial agreement with other reports (De La Cruz et al. 2001). It is important to confirm whether this improvement in bunch weight is repeatable, because an additional yield of 2.7 Mt might be obtained, based on exportable banana production per hectare at 1.05 boxes/bunch, 1700 effective production units per hectare, a ratooning rate (number of bunches per production unit per year) of 1.4, and 25% rejected bananas. The additional yield may cover the extra expense incurred by manual weed control. A similar effect on yield was referred to by Bhattacharyya and Madhava (1987), who found that banana and sugarcane leaf trash, used as a soil cover, enhanced banana yield. This approach to weed management is being undertaken in some Colombian banana plantations (Pinilla and García 2002).

Mulched banana plantations may result in more sustained yield compared with unmul-ched areas. Although manual weed control appears to be beneficial for banana yield, it is a labour-intensive activity and requires a well-planned program to assure its positive effects. As with other crops, the effects of mulch on banana growth and yield may vary with physical and chemical soil properties, microclimate, and other factors. Therefore,


its introduction into a farm needs prior evaluation and a combined strategy of manual weed control and mulching is highly encouraged.

A deep rooting system should allow banana plants to better resist drought. Burgess et al. (1998) demonstrated a redistribution of soil water from deeper in the profile to dry surface horizons by the root system. This should maintain root viability, facilitating root growth during dry periods. Araya et al. (1999), working with plants with differences in vigour, found about 11% of the total root system below 60 cm depth in regular and good vigour plants. Probably, the deeper roots provided the banana plants with better resis-tance to drought in the months of February, March and April when precipitation was low. Additionally, a deep root system could be advantageous in the use of soil resour-ces. It is possible to improve the depth of rooting by soil preparation in replanting either by sub-soiling, ploughing or disking according to soil requirements. Consequently, in addition to soil preparation, domes are being increasingly used in banana replanting. Pérez (1999) reported greater root mass at greater depth in plants cultivated on domes compared with those grown on flat surfaces. Robinson and Alberts (1989), working with ‘Williams’ banana, found most water uptake in the upper layers of the soil; up to 80% of water being taken from the upper 30 cm. In places where irrigation is used, more work is needed to assess available water and the advantages of using a specific irrigation system because water application needs to account for the amount of availa-ble water in the entire rooting zone, which depends on the rooting pattern and rooting depth.

Fertilizer application must take into consideration the distribution of roots across the soil surface. All root types were recorded at the three horizontal distances from the pseudostem. The proportion of thin and fine roots together varied between 33 and 50% of the total root system and depended on plant age, vigour and soil condition. According to Swennen et al. (1986 and 1988) banana feeder roots are those that are less than 5 mm diameter, and in Musa acuminata triploids (AAA), cultivated in hydroponics, they made up as much as 77% of the root system.

Mohan and Madhava (1988) found that plants do not draw nutrients equally from all zones in the soil through which their roots penetrated. For bananas, they stated that late-ral roots bearing hairs are mainly responsible for nutrient and water uptake. These roots are distributed away from the corm and hence the placement of fertilizer should also be away from the plant (Summerville 1939). Studies of Mohan and Madhava (1988) revealed that the active zone of nutrient uptake for 2 month-old bananas was within 30 cm of the plant. However, in 5 month-old plants no large differences in the active zone were found up to 60 cm from the corm. On the basis of the findings presented here, where thin and fine roots were detected at 0-30, 30-60 and 60-90 cm from the stem, and those of Mohan and Madhava (1988) it appears advisable to apply fertilizer, taking plant development into consideration, doing broadcast applications between the plant base and up to 60-90 cm from the stem, since roots spread well. This recommendation agrees with that of Mohan and Madhava (1988), Keshava and Iyengar (1990 and 1997) and Araya et al. (1998). Broadcast applications should reduce the common problem of soil acidity and accumulation of specific nutrients in a small band or area affecting nutrient uptake and yield. For soil nutrient analysis, the sample must correspond with the active root zone, where fertilizer was applied.


Based on the spatial horizontal root distribution, the samples for nematode studies could be taken at any distance from the stem. Nevertheless, according to the spatial root nematode distribution detected by Araya et al. (1999) and Araya and De Waele (2002), samples for monitoring nematode populations should be taken within a 30 cm radius of the plant base and down to 30 cm depth from the soil surface. This allows for good sample sizes to be obtained in smaller areas (15 cm long x 15 cm wide x 30 cm depth) because of high root density. The suggested sampling point is consistent with the recommendation of Cabrales (1995) and Pattison et al. (1997).

Lassoudière (1978), Robinson (1996) and Serrano and Marín (1998) reported positive correlations between bunch weight or yield of dessert bananas and root mass. Calvo and Araya (2001) found good yield when root mass in nematode samplings was high. Attempts to relate the crop performance with the extent of rooting are desirable because it is known that the benefits derivable from a given investment in root dry weight depend fundamentally on root system architecture (Fitter 1996).

ReferencesAguilar E.A., D.W. Turner & K. Sivasithamparam. 1999. Aerenchyma formation in roots of four banana (Musa spp.)

cultivars. Scientia Horticulturae 80:57-72.

Aguilar E.A., D.W. Turner, D.J. Gibbs, W. Armstrong & K. Sivasithamparam. 2003. Oxygen distribution and movement, respiration and nutrient loading in banana roots (Musa spp. L.) subjected to aerated and oxygen-depleted environ-ments. Plant and Soil 253:91-102.

Aguilar V.A. 2002. Identificación de hongos asociados al deterioro radicular del banano (Musa AAA) y su relación con el daño causado por el nematodo barrenador (Radopholus similis, Cobb) en fincas de la vertiente del Caribe de Costa Rica. Thesis Agr. Ing. ITCR, San Carlos, Costa Rica. 57pp.

Araya M. & D. De Waele. 2002. Effect of soil type on the spatial distribution of Radopholus similis on banana roots. International Journal of Nematology 12:137-144.

Araya M. & F. Blanco. 2001. Changes in the stratification and spatial distribution of the banana (Musa AAA cv. ‘Grande Naine’) root system of poor, regular, and good developed plants. Journal of Plant Nutrition 24:1679-1693.

Araya M., A. Vargas & A. Cheves, A. 1998. Changes in distribution of roots of banana (Musa AAA cv. ‘Valery’) with plant height, distance from the pseudostem, and soil depth. Journal of Horticultural Science & Biotechnology 73:437-440.

Araya M., A. Vargas & A. Cheves. 1999. Nematode distribution in roots of banana (Musa AAA cv. ‘Valery’) in relation to plant height, distance from the pseudostem and soil depth. Nematology 1:711-716.

Atkinson D. 1989. Root growth and activity: current performance and future potential. Annals of Applied Biology 22:1-13.

Avilan R.L., L. Meneses, R. Sucre, C. Orta & O. Sangle. 1979. Distribución del sistema radical del banano ‘Pigneo Gigante’ bajo cuatro sistemas de manejo de suelos. Agronomía Tropical 29:299-312.

Avilan R.L., R. Meneses & R.E. Sucre. 1982. Distribución radical del banano bajo diferentes sistemas de manejo de suelos. Fruits 37:103-109.

Azcón A.C., M.C. Jaizme & C. Calvet. 2002. The contribution of arbuscular mycorrhizal fungi to the control of soil-borne plant pathogens. Pp. 187-197 in Mycorrhizal Technology in Agriculture (S. Gianinazzi, H. Schuepp, J.M. Barea & K. Haselwandter, eds). Birkhauser Verlag/Switzerland.

Barber S.A. 1992. Roots- past, present and future. Pp. 19-29 in Proceedings of plant nutrition conference. Addressing the physical/chemical/biological/environmental interactions relating to root growth and plant nutrition. July 8-12, 1992 (H.F. Reetz, ed.). Champaign, Illinois, USA.

Batlle V.A. & Pérez V.L. 2002. Estudio de las especies de hongos endofíticos asociados a la necrosis de las raíces de bananos en plantaciones de bananos y plátanos de Cuba. InfoMusa 11(1):23-25.

Bennie A.T.P. 1996. Growth and mechanical impedance. Pp. 453-470 in Plant roots, the hidden half (Y. Waisel, A. Eshel and U. Kafkafi, eds). Marcel Dekker, Inc., New York, USA.



Bhattacharyya R.K. & R.V.N. Madhava. 1986. Management. Effect of soil covers on soil temperature in a banana plantation. Banana Newsletter 9:8-9.

Bhattacharyya R.K. & R.V.N. Madhava. 1988a. Leaf production of ‘Robusta’ banana as influenced by soil covers and soil moisture regimes. Banana Newsletter 11:4-5.

Bhattacharyya R.K. & R.V.N. Madhava. 1988b. Root cation exchange capacity (CEC) of ‘Robusta’ banana as influ-enced by soil covers and soil moisture regimes. Banana Newsletter 11:6-7.

Bhattacharyya R.K. & R.V.N. Madhava. 1989. Climate and irrigation. Shooting-harvest Interval of ‘Robusta’ banana as influenced by soil covers and soil moisture regimes. Banana Newsletter 12:22-23.

Bhattacharyya R.K. & R.V.N. Madhava. 1990. Water relations. Leaf relative water content of ‘Robusta’ banana as influenced by soil covers and soil moisture regimes. Banana Newsletter 13:17-18.

Bhattacharyya R.K. & R.V.N. Madhava. 1992. Root penetration in depth of cv. ‘Robusta’ banana (AAA) as influenced by soil covers and soil moisture regimes. Banana Newsletter 15:18-19.

Bhattacharyya R.K. & R.V.N. Madhava. 1987. Yield of ‘Robusta’ banana as influenced by soil covers and soil moisture regimes. Banana Newsletter 10:9-10.

Blal B., C. Morel, V. Gianinazzi-Pearson, J.C. Fardeau & S. Gianinazzi. 1990. Influence of vesicular-arbuscular mycor-rhizae on phosphate fertilizer efficiency in two tropical acid soils planted with micropropagated oil palm (Elaeis guineensis jacq.). Biology and Fertility of Soils 9:43-48.

Blomme G. & R. Ortiz. 1996. Preliminary assessment of root systems in Musa. MusAfrica 10:16.

Blomme G. 2000. The interdependence of root and shoot development in banana (Musa spp.) under field conditions and the influence of different biophysical factors on this relationship. PhD Thesis. Katholieke Universiteit Leuven, Belgium. 183pp.

Blomme G., R. Swennen & A. Tenkouano A. 2003. Assessment of genotypic variation in the root architecture of Musa spp. under field conditions. InfoMusa 12(1):24-29.

Blomme G., R. Swennen & A. Tenkouano. 2000. Assessment of variability in the root system characteristics of banana (Musa spp.) according to genome group and ploidy level. InfoMusa 9(2):4-7.

Blomme G., R. Swennen & A. Tenkouano. 2002. The effect of soil bulk density on root and overall plant development in six banana varieties. InfoMusa 11(2):38-40.

Blomme G., R. Swennen, A. Tenkouano & R. Ortiz. 2001. Estimation of root development from shoot traits in plantain and banana (Musa spp.). InfoMusa 10(1):15-17.

Blomme G., X. Draye, G. Rufyikiri, S. Declerck, D. De Waele, A. Tenkouano & R. Swennen. 2000. Progress in under-standing the roots of Musa spp. Pp. 14-19 in INIBAP Annual Report 1999. INIBAP, Montpellier, France.

Borges A.L., O.A.M. Gómes & L.S. Souza. 1999. Solos, nutricao e adubacao. Pp. 197-260 in A cultura da banana. Aspectos técnicos, socioeconomicos e agroindustriais (E.J. Aves, ed.). Empresa Brasileira de Pesquisa Agropecuária. Centro Nacional de Pesquisa de Mandioca e Fruticultura Tropical. Ministério da Agricultura e do Abastecimento. Brasilia, DF.

Burgess S.S.O., M.A. Adams, N.C. Turner & C.K. Ong. 1998. The redistribution of soil water by tree root systems. Oecologia 115:306-311.

Cabrales M.L. 1995. Determinación del sitio de muestreo de raíces en estudios nematológicos del banano. Fitopatología Colombiana 19:72-77.

Calvo C. & M. Araya. 2001. Cantidad de raíces de banano en los diez cantones productores de Costa Rica. Corbana 27:47-64.

Champion J. & D. Sioussaram. 1970. L’enracinement du bananier dans les conditions de la station de Neufcháteau (Guadaloupe). Fruits 25 847-859.

Charlton W.A. 1996. Lateral root initiation. Pp. 149-173 in Plant roots: The hidden half (Y. Waisel, A. Eshel & U. Kafkafi, eds). Marcel Dekker, Inc., New York, USA.

De La Cruz R., C.E. Rojas, H.L. Lobón & C. Burgos. 2001. El papel de las malezas en la reducción de la lixiviación de nutrimentos en cultivos de enano en el trópico húmedo. Manejo Integrado de Plagas 62:29-37.

De Moura A.R.B., F.O.S. De Araujo, J.C. Taveres & M. De Oliveira. 1986. Estimativa da densidade radicular da bananeira ‘Nanica’ submetida a um sistema nao convencional de irrigacao. Pp. 83-86 in Congreso Brasileño de Fruticultura. (8, Brasilia,1986). Anais, Brasil, Brasilia S.B.F.


De Waele D. 2000. Root-knot nematodes. Pp. 307-314 in Diseases of Banana, Abacá and Enset (D.R. Jones, ed.). CABI Publishing, Wallingford, UK.

De Waele D., S.R. Gowen, M. Tessera & A.J. Quimio. 2000. Other nematodes. Pp. 314-323 in Diseases of Banana, Abacá and Enset (D.R. Jones, ed.). CABI Publishing, Wallingford, UK.

Delvaux B. 1995. Soils. Pp. 230-257 in Bananas and Plantains (S. Gowen, ed.). Chapman & Hall, London, UK.

Dorel M. 2002. Physical properties of the volcanic soils of Guadeloupe used for banana growing: influence of cultural practices and their effects on rooting. InfoMusa 11(2):48.

Draye X., B. Delvaux & R. Swennen. 1999. Distribution of lateral root primordia in root tips of Musa. Annals of Botany 84:393-400.

Drew M.C. & L.H. Stolzy. 1996. Growth under oxygen stress. Pp. 397-414 in Plant Roots: The Hidden Half (Y. Waisel, A. Eshel & U. Kafkafi, eds.). Marcel Dekker, Inc., New York, USA.

Elsen A., S. Declerck & D. De Waele. 2002. Effect of three arbuscular mycorrhizal fungi on root-knot nematode (Meloidogyne spp.) infection of Musa. InfoMusa 11(1):21-23.

Esau K. 1965. Plant Anatomy. (2nd ed.). John Wiley & Sons, New York, USA.

Fitter A. 1996. Characteristics and functions of root systems. Pp. 1-20 in Plant roots: The hidden half (Y. Waisel, A. Eshel & U. Kafkafi, eds.). Marcel Dekker, Inc., New York, USA.

Gao S., W.L. Pan & R.T. Koenig. 1998. Integrated root system age in relation to plant nutrient uptake activity. Agronomy Journal 90:505-510.

Ghavami M. 1976. Banana plant response to water table levels. Transactions of the American Society of Agricultural Engineers 16:675-677.

Girija V.K. & S.K. Nair. 1988. Incidence of VAM in banana varieties. Indian Journal of Microbiology 28:294-295.

Godefroy J. 1969. Développement des racines du bananier dans divers sols: relation avec la fertilité. Fruits 24:101-104.

Goodman A.M. & A.R. Ennos. 1999. The effects of soil bulk density on the morphology and anchorage mechanics of the root systems of sunflower and maize. Annals of Botany 83:293-302.

Gousseland J. 1983. Etude de l’énracinement et de l’emission racinaire du bananier ‘Giant Cavendish’ (Musa acuminata AAA, sous-groupe Cavendish) dans les andosols de la Guadeloupe. Fruits 38:611-623.

Gowen S. & P. Quénéhervé. 1990. Nematode parasites of bananas, plantains and abaca. Pp. 431-460 in Plant para-sitic nematodes in subtropical and tropical agriculture (M. Luc, R. Sikora & J. Bridge, eds). CAB International, Wallingford, UK.

Gowen S.R. 1995. Pests. Pp. 382-402 in Bananas and Plantains (S. Gowen, ed.). Chapman & Hall, London, UK.

Gowen S.R. 2000a. Root-lesion nematodes. Pp. 303-306 in Diseases of Banana, Abacá and Enset (D.R. Jones, ed.). CABI Publishing, Wallingford, UK.

Gowen S.R. 2000b. Spiral nematode. Pp. 306-307 in Diseases of Banana, Abacá and Enset (D.R. Jones, ed.). CABI Publishing, Wallingford, UK.

Hegde D.M. & K. Srinivas. 1989. Effect of soil moisture stress on fruit growth and nutrient accumulation in banana Cultivar ‘Robusta’. Fruits 44:135-138.

Irizarry H., J.V. Chandler & S. Silva. 1981. Root distribution of plantains growing on five soil types. Journal of Agriculture of the University of Puerto Rico 65:29-34.

Irizarry H., S. Silva & J.V. Chandler. 1980. Effect of water table level on yield and root system of plantains. Journal of Agriculture of University of Puerto Rico 64: 33-36.

Itai C., & H Birnbaum. 1996. Synthesis of plant growth regulators by roots. Pp. 273-284. in Plant roots: The hidden half. (Y. Waisel, A. Eshel, & U. Kafkafi, eds). Marcel Dekker, Inc. New York, USA.

Iyer R., H. Moosa & S.R. Kalpana. 1988. Vesicular-arbuscular mycorrhizal association in banana. Current Science 57:153-155.

Jaizme V.M.C., D.M. Esquivel, D.P. Tenoury & R.A.S. Rodríguez. 2002. Efectos de la micorrización sobre el desarrollo de dos cultivares de platanera micropropagada. InfoMusa 11(1):25-28.

John S.T.V. 1988. Prospects for application of vesicular-abuscular mycorrhizae in the culture of tropical palms. Advances in Economic Botany, 6: 50-55.


Keshava M.S.V. & B.R.V. Iyengar. 1990. Effect of time and placement on fertilizer P uptake by ‘Robusta’ banana. Journal of Nuclear Agriculture and Biology 19:247-269.

Keshava M.S.V. & B.R.V. Iyengar. 1997. Root distribution and morphology in some banana (Musa x paradisiaca) varieties. Indian Journal of Agricultural Sciences 67:495-499.

Klepper B. 1992. Roots: past, present, and future. Pp. 7-18 in Proceedings root of plant nutrition conference. Addressing the physical/chemical/biological/environmental interactions relating to root growth and plant nutrition. July 8-12, 1992. (H.F. Reetz, ed.). Champaign, Illinois. USA.

Kobenan K., S. Aké & D. Koné. 1997. Study of the reactions of the roots of banana Musa (AAA) cv ‘Grande Naine’ to artificial lesions. InfoMusa 6(1):21-23.

Lahav E. & D.W. Turner. 1992. Fertilización del banano para rendimientos altos. Instituto Internacional de la Potasa. Boletín No 7-IPI, Quito, Ecuador.

Lassoudière A. 1971. La croissance des racines du bananier. Fruits 26:501-512.

Lassoudière A. 1978. Quelques aspects de la croissance et du développement du bananier ‘Poyo’ en Cote d’Ivoire. II Le système radical. Fruits 33:314-338.

Lavigne C. 1987. Contribution a I´étude du système racinaire du bananier. Mise au point de rhizotrons et premiers résultats. Fruits 42:265-271.

Loridat Ph. & J. Ganry. 1991. Mise en évidence d´une interaction nématode-champignon (Radopholus similis/Cylindrocladium), comme composante du parasitisme tellurique du bananier en culture industrielle aux Antilles. Pp. 283-303 in IX Reunión ACORBAT. Mérida, Venezuela, 1989.

Martinez R. & E. Ortega. 1988. Distribución vertical del sistema radical y consumo por capas en dos clones de plátano fruta (Musa acuminata). Cienc. Tec. Agric. Riego y Drenaje 7: 33-43.

Mathew V., M. Arvindakshan, P.K. Valsalakumari & N.K. Parameswaran. 1987. Root distribution of banana cultivars under rainfed condition. South Indian Horticulture 35:334-338.

McCully M.E. 1999. Roots in soil: unearthing the complexities of roots and their rhizospheres. Annual Review of Plant Physiology and Plant Molecular Biology 50:695-718.

McMichael B.L. & J.J. Burke. 1996. Temperature effects on root growth. Pp. 383-396 in Plant roots the hidden half. (Y. Waisel, A. Eshel & U. Kafkafi, eds.). Marcel Dekker, Inc. New York, USA.

Mohan N. & V.N.M. Rao. 1984. The effect of plant density on the banana root system. South Indian Horticulture 32:254-257.

Mohan N.K. & R.V.N. Madhava. 1988. Tracer studies to determine the active root zone in banana. Pp. 79-83 in Memoria 1986 de la IV reunión sobre agrofisiología del banano. (J.A. Guzmán & R. Romero eds). Asociación Bananera Nacional (ASBANA), Costa Rica.

Obiefuna J.C. 1991. The effect of crop residue mulches on the yield and production pattern of plantain (Musa AAB). Biological Agriculture and Horticulture 8:71-80.

Parvatha R.P., M.S. Rao & M. Nagesh. 2000. Integrated management of burrowing nematode (Radopholus similis) using endomycorrhiza (Glomus mosseae) and oil cakes. Pp. 344-348 in Banana Improvement, Production & Utilization. (H.P. Singh & K.L. Chadha, eds.). AIPUB / NHB / INIBAP, India.

Pattison T., J. Stanton & S. Lindsay. 1997. Burrowing nematode management, for sustainable management of burrowing nematode in bananas. Queensland Department of Primary Industries, Information series Ql97034. 37pp.

Pérez S.L. 1999. Renovación de plantaciones de banano y su relación con la presencia de raíces sanas y producciones altas. Corbana 25:211-216.

Pezeshki S.R., R.D. DeLaune & P.H. Anderson. 1999. Effect of flooding on elemental uptake and biomass allocation in seedlings of three bottomland tree species. Journal of Plant Nutrition 22:1481-1494.

Pinilla G.C.H. & C.J.M. García. 2002. Manejo integrado de malezas en plantaciones de banano. C.I. UNIBAN, S.A. 2nd ed. 296pp.

Pinochet J.G. & R.H. Stover. 1980. Fungi in lesions caused by burrowing nematodes on bananas and their root and rhizome rooting potential. Tropical Agriculture 57:227-232.

Pocasangre L., R.A. Sikora, V. Vilich & R.P. Shuster. 2000. Survey of banana endophytic fungi from Central America and screening for biological control of Radopholus similis. Acta Horticulturae 531:283-289.

Price N.S. 1995. Banana morphology - part I: roots and rhizomes. Pp. 179-189 in Bananas and Plantains. (S. Gowen, ed.). Chapman & Hall, London, United Kingdom.


Riopel J.L. 1966. The distribution of lateral roots in Musa acuminata ‘Gross Michel’. American Journal of Botany 53: 403-407.

Robinson J.C. 1987. Root growth characteristics in banana. ITSC Information Bulletin No 183: 7-9.

Robinson J.C. 1996. Bananas and Plantains. CAB International, Wallingford, UK.

Robinson J.C. & A.J. Alberts. 1989. Seasonal variation in the crop water-use coefficient of banana (cultivar ‘Williams’) in the subtropics. Scientia Horticulturae 40:212-225.

Robinson J.C., A.J. Alberts & R.E. Reynolds. 1989. Irrigating bananas in the subtropics. Banana Newsletter 12:23-25.

Robinson J.C., A.J. Alberts & R.E. Reynolds. 1991. Water relations, C/N ratio in leaves of ‘Robusta’ banana as influen-ced by soil covers and soil moisture regimes. Banana Newsletter 14:15-16.

Ruhigwa B.A., M.P. Gichuru, D.S.C. Spencer & R.L. Swennen. 1995b. Economic analysis of cut-and-carry and alley cropping systems of mulch production for plantains in Southeastern Nigeria. IITA Research 11:11-14.

Ruhigwa B.A., M.P. Gichuru, R.L. Swennen & N.M. Tariah. 1995a. Plantain production in an alley cropping system on an Ultisol in Southeastern Nigeria. Pp. 268-277 in Alley Farming Research and Development. Proceedings of the International Conference on Alley Farming. (B.T. Kang, A.O. Osiname, A. Larbi, eds). African Book Builders Ltd. Ibadan, Nigeria.

Salau O.A., O.A. Opara-Nadi & R.L. Swennen. 1992. Effects of mulching on soil properties, growth and yield of plan-tain on a Tropical Ultisol in Southeastern Nigeria. Soil and Tillage Research 23:79-93.

Sandoval J.A. & L. Müller. 1999. Anatomía y morfología de la planta de banano (Musa AAA). Corbana 24:43-59.

Sarah J.L. 2000. Burrowing nematode. Pp. 295-303 in Diseases of Banana, Abacá and Enset. (D.R. Jones, ed.). CABI Publishing, Wallingford, UK.


Sikora R.A. & E. Schlosser. 1973. Nematode and fungi associated with root systems of bananas in a state of decline in Lebanon. Plant Disease Reporter 57:615-618.

Sioussaram D. 1968. Observations préliminaires sur l’enracinement des bananiers dans les sols de la station de Neufchâteau, Guadeloupe. Fruits 17:473-479.

Skutch A.F. 1932. Anatomy of the axis of the banana. Botanical Gazette 43:233-258.

Soto B.M. & E. Ruiz. 1992. Descripción botánica. Pp. 21-103 in Bananos: Cultivo y Comercialización. (M. Soto ed.). Lil, S.A. Costa Rica.

Stover R.H. & N.W. Simmonds. 1987. Bananas. (3rd ed.). Longman, Harlow, United Kingdom.

Stover R.H. 1966. Fungi associated with nematode and non-nematode lesions on banana roots. Canadian Journal of Botany 44:1703-1710.

Stover R.H. 1972. Banana, Plantain and Abaca Diseases. Commonwealth Mycological Institute, Kew, UK.

Summerville W.A.T. 1939. Root distribution in banana. Queensland Agricultural Journal 52:376-392.

Swennen R. & Wilson G.F. 1985. Flemingia congesta as a life cover plant for plantain. Banana Newsletter 8: 17-18..


Swennen R., G.F. Wilson & D. Decoene. 1988. Priorities for future research on the root system and corm in plantains and bananas in relation with nematodes and the banana weevil. Pp. 91-96 in Nematodes and the Borer Weevil in Bananas. Proceedings of a workshop, Bujumbura, Burundi, 1987. Montpellier, INIBAP.

Talwana H.A.L., P.R. Speijer, C.S. Gold, R.L. Swennen, & D. De Waele. 2003. A comparison of the effects of the nema-todes Radopholus similis and Pratylenchus goodeyi on growth, root health and yield of an East African highland cooking banana (Musa AAA-group). International Journal of Pest Management 49:199-204.

Umesh K.C., K. Krishnappa & D.J. Bagyaraj. 1988. Interaction of burrowing nematode, Radopholus similis (Cobb, 1893) Thorne, 1949 and VA mycorrhiza, Glomus fasciculatum (Thaxt.) Gerd & Trappe in banana (Musa acuminata Colla.) Indian Journal of Nematology 18:6-11.

Vargas R. & C.L. Flores. 1995. Retribución nutricional de los residuos de hojas, venas de hojas, pseudotallo y pinzote de banano (Musa AAA) en fincas de diferentes edades de cultivo. Corbana. 20:33-47.

Villalobos C., F.A. Blanco & R. Vargas. 2001. Efecto de la edad de plantación sobre el potencial de inóculo y la infec-tividad de las poblaciones nativas de hongos micorrícicos arbusculares (MA) en plantaciones de banano (Musa AAA cv. Gran Enano). Corbana 27:35-46.


Waisel Y., & A. Eshel. 1992. Differences in ion uptake among roots of various types. Journal of Plant Nutrition 15:945-958.

Weert R.V.D. 1974. Influence of mechanical forest clearing on soil conditions and the resulting effects on root growth. Tropical Agriculture 51:325-331.

Wiersum L.K. 1957. The relationship of the size and structural rigidity of pores to their penetration by roots. Plant and Soil 9:75-85.

Zobel R.W. 1992. Soil environment constraints to root growth. Advances in Soil Science 19:27-51.

C.A. Gauggel et al. 105

3

Root physiology

Fisiología de la raíz

The problem of banana root deterioration and its impact on production106 Factors affecting the physiology of the banana root system C.A. Gauggel et al. 107D.W. Turner

Factors affecting the physiology of the banana root systemD W Turner1

AbstractThe roots of bananas and plantains absorb nutrients and water from the medium in which they grow and transport these to the shoot. The absorption of nutrients and their loading into the stele for transport to the shoots requires energy and is very sensitive to oxygen deficiency, a component of waterlogging. In contrast, the gradient in water potential together with the conductance of the pathways of water flow, which can be influenced by the plant and its environment, determines the direction and volume of water flow. No metabolic energy is needed for this process. Under drought, roots produce signals that close stomata, allowing the banana to remain highly hydrated, but reduce carbon assimilation and yield. In addition, roots are the ‘home’ for micro-organisms that influence the physiology of the roots. A way forward is to integrate our knowledge into quantitative models that evaluate the most sensitive features of the root system in relation to its functions. This will point the way to improved management practices and genetic manipulation. A large gap in our knowledge is the genetic variation in physiological responses of roots to their environment, especially the rhizosphere.

Resumen - Factores que afectan la fisiología del sistema radical del bananoLas raíces de los bananos y los plátanos absorben nutrientes y agua del medio en que crecen y lo transportan a los brotes en crecimiento de la planta. La absorción de nutrientes y su carga en el cilindro vascular para su transporte a los brotes en crecimiento, requieren energía y es muy sensitivo a la falta de oxigeno, una característica del estancamiento de agua. En contraste, la gradiente en el potencial del agua junto con la conductividad del camino del flujo del agua, que puede ser influenciado por el ambiente, determina la dirección y el volumen del flujo del agua. En condiciones de sequía, las raíces producen señales que cierran los estomas, permitiendo al banano permanecer altamente hidratado, pero reduce la asimilación del carbono y la producción. Además, las raíces son “la casa” de los microorganismos que influencian la fisiología radical. Un camino hacia delante es él poder integrar nuestros conocimientos en modelos cuantitativos que evalúen las características más sensibles del sistema radical con relación a sus funciones. Esto señalara la ruta para mejorar las practicas de manejo y manipuleo genético. Un gran vació en nuestro conocimiento es la variación genética en respuestas fisiológicas de las raíces en su ambiente, especialmente en la rizosfera.

IntroductionPhysiology is about how an organism works. To manage a crop requires knowledge of how that crop works at different levels. The manager is most attracted to knowledge at the whole plant level but this needs to be supported by knowledge of plant organs, their tissues and the cells that make up the tissues. Price (1995) pointed out that for bananas and plantains, “research on root physiology, metabolism and interactions with the soil and soil microorganisms is in its infancy” and Delvaux (1995) concluded that “our understanding of soil-root-microbial interactions is still embryonic”. Therefore, investment in exploring the physiology of banana root systems should reward us with new insights and approaches to management.

1 School of Plant Biology (M084). Faculty of Natural and Agricultural Sciences. The University of Western Australia. 35 Stirling Hwy, Crawley WA 6009. Australia. e-mail: [email protected]


Draye (2002) has reviewed the banana root system with emphasis on architecture and genetics. My purpose is to draw some links between the architectural, structural and anatomical features of banana roots and their physiology. Superimposed on these links are the environmental, edaphic and biological factors that affect roots.

How well does the banana root system work?Banana cultivars can be highly productive and in commercial plantations can produce 80 to 100 t-1 ha-1 year-1 of fresh fruit. This approximates 16 to 20 tonnes dry matter and puts bananas in the top league among plants for productivity. Thus, the root system of bananas with its current architecture and function, and with high management inputs, can support high production. However, most banana plantations yield less because the plants, including the root system, are constrained by genetic, environmental, edaphic and biological factors.

Root length density (Ld, cm/cm3) and specific root length (Lw, m/g) are quantitative features of the architecture of root systems. Both may be subject to genetic manipu-lation within Musa. Root length density quantifies the capacity of the root system to explore soil volume. A high Ld means that the roots absorb more of the nutrients in a volume of soil, especially those nutrients that diffuse to the root surface. Banana and plantain roots have an Ld of about 1 cm/cm3 (Irizarry et al. 1981) that is similar to the root systems of trees. On the other hand, herbaceous species have an Ld, in the surface layers of soil, of 4 to 50 cm/cm3.

Specific root length quantifies the length of root per unit of carbon invested in the root system. In bananas, the main, secondary and tertiary nodal roots and the main and secondary lateral roots have a unique Lw. The banana invests carbon in nodal roots, that arise from the rhizome (corm) and which have a much longer life than the lateral roots which absorb most of the water and nutrients. For mature plants of ‘Williams’ growing in sand culture the Lw varies from 0.4 m/g for the main nodal roots to 150 m/g for the secondary lateral roots (Turner and Barkus 1981). The partitioning of dry matter between nodal and lateral roots is about 50:50 but this is likely to change during plant development. In the example above, the allocation of 1 g of dry matter to main nodal roots will extend the root by 0.4 m but allocation of the same amount to secondary lateral roots will provide a total 150 m of roots. How Lw changes with genotype, plant development, climatic and edaphic conditions needs to be determined. The values of Lw for bananas can be compared with 100 to 200 m/g for the whole root system of cereals and grasses (Milthorpe and Moorby 1979). This difference in root architecture between bananas and grasses may contribute to the competitive success of grasses when they grow amongst bananas.

Root growth rates have been measured in some controlled environment experiments and in the plantation. The dry matter of the root system of ‘Williams’ follows an opti-mum curve with temperature with little growth below 13˚C or above 35˚C and an optimum of about 22˚C (Turner and Lahav 1983). For the rate of root elongation, the response to temperature may be exponential from 10 to 26˚C (Robinson and Alberts 1989) or show an optimum at about 23˚C, depending on the assumptions made in the analysis of Robinson and Alberts’ data (Turner 1994). The impact of soil water deficits on root growth needs to be determined.


Oxygen supplyRoots are living entities and need oxygen for their growth and function (Armstrong 1979). Oxygen deficiency is a component of waterlogging caused by poor internal drainage in soils. Delvaux (1995) lists eight soil orders on which bananas are grown across the world and for six orders, drainage is a key management practice. Therefore, knowledge about how banana and plantain roots respond to oxygen deficiency is important for management.

Roots adapt to oxygen deficiency to various degrees by changing their anatomy and metabolism. Banana roots develop lysigenous aerenchyma in the mid cortex from about 50 mm behind the root tip, even in aerated media (Aguilar et al. 1999). Among cultivars, genomic constitution appears not to influence aerenchyma formation. Aerenchyma, provided it is continuous, increases gas flow axially along the root and, in waterlogged conditions, enables oxygen to move from the shoot to the roots (Aguilar et al. 2003). The stele of roots of ‘Williams’ respires (on a volume basis) six times faster than the cortex. This is about 50% higher than in maize, for example, and contributes to a rapid depletion of oxygen concentration in the stele of roots experiencing waterlogging. Even in fully aerated solution culture, oxygen concentration can be halved from the bulk medium across the root hair zone to the root surface. Further reductions inside the root mean that even a small drop of oxygen concentration, say by 3 kPa in the medium, is likely to cause the stele to become anoxic and reduce nutrient loading into the stele. Under laboratory conditions, net nutrient loading begins to fall soon after the imposition of hypoxia and after 4 hours is 36% of its original rate. Lateral roots consume oxygen that flows longitudinally from the shoot (Aguilar et al. 2003). They reduce the amount of oxygen that can then flow to the apex but are essential for the root to explore large volumes of new soil. The density of the short secondary lateral roots is within the range to classify them as cluster roots, however we should be careful in assuming they are physiologically similar until data are available on their function. Lack of oxygen kills root tips (Aguilar et al. 2003) with subsequent branching behind the dead apex. This is commonly seen in the plantation and may indicate an earlier waterlogging event.

Roots as a “home” for microorganismsRoots provide an environment for microorganisms, such as fungi and nematodes. The aerenchyma provides a suitable habitat for Fusarium (Aguilar et al. 2000a). Indeed, hyphal connections between the aerenchyma and the stele may be necessary for Fusarium, as the hyphae in the aerenchyma can absorb oxygen for transport to the stele, which may be anoxic. In addition, the hyphae in the stele can absorb nutrients for transport to the hyphae in the aerenchyma. Beckman et al. (1961) proposed that physical barriers and physiological responses are essential for resistance to Fusarium. Temporary blockage of the fungus at the vessel endings in the stele, for example, would provide time for physiological responses to be effective. However, aerenchyma, if present, may provide a by-pass for the fungal hyphae around barriers in the stele. Mycorrhizae increase the acquisition of phosphorus and potassium in banana coming from micropropagation (Declerck et al. 1994) and there is an interaction between the species of mycorrhizal fungi and cultivar (Knight 1988). Fungi and nematodes obtain nourishment from the root system and this represents a carbon cost for the plant.


Microorganisms alter the physiology of the roots they invade. Fusarium oxysporum f. sp. cubense (Foc) increases the activities of enzymes involved in phenol metabolism in roots and so does exposure to hypoxia, more so in cultivars resistant to Foc (Aguilar et al. 2000b). These interactions are fascinating but complex and make it difficult to predict what might happen in the field.

Nutrient uptake by root systemsThe efficiency with which the root system transfers nutrients from the soil to the plant is a gross measure of root function. The apparent root transfer coefficient, α, proposed by Nye and Tinker (1969), evaluates this efficiency and it is suitable for longer term (weeks or months) studies. The plant uptake rate for a nutrient, Rp, is influenced by the root dry weight, Wr, the concentration of the nutrient at the root surface, C0 and α. This approach includes growth and nutrient components and is calculated from Nye and Tinker (1969):

α = (W/Wr)(Cm/C0)(RW + RC) 1.

Where W is plant weight, Cm is the nutrient concentration in the plant, RW is the relative growth rate and RC is the relative change in nutrient concentration over time. In equa-tion 1, W/Wr and RW are growth components and the nutrient components are Cm/C0 and RC. To increase plant uptake rate of a nutrient we increase either root weight (e.g. improve soil conditions for roots), or the concentration of nutrient at the root surface (e.g. add fertilizer). The α can be influenced by environment and nutrient supply. For example, Turner and Lahav (1983) studied the effect of temperature on the growth of ‘Williams’ banana plants in the vegetative stage in controlled environments over 12 weeks. As temperature increased from 25/18˚C (day/night) to 33/26˚C, the root dry weight, Wr, fell by almost two thirds. This could be expected to reduce K uptake. However, temperature increased the apparent root transfer coefficient α of K so that it more than compensated for the reduced Wr and plant uptake rate was increased by 50% (Turner and Lahav 1985).

Earlier, Turner and Barkus (1981) evaluated the effect of K supply on α in a sand cul-ture experiment lasting three crop cycles. Low K supply reduced yield and K deficiency reduced the effectiveness of the root system, α, in absorbing N, P, Ca, Mn, Na, and Zn but had no effect on the α of K, Mg or Cu. These differences were caused by the mar-ked effect of K on growth, and in the case of K, Mg and Cu there were compensatory changes in the concentration components.

In summary, the environmental factor, temperature, generally increases root system effectiveness for all nutrients from 13 to 30˚C, overriding some of the negative effects on root growth above 22˚C. Potassium deficiency reduces the effectiveness of the root system for absorbing most nutrients, mainly through its effect on reduced growth. However, a deficiency of K does not reduce the effectiveness of the root system in absorbing the K that is available, despite the reduction in growth and yield.

Water uptake and banana rootsRoots absorb and conduct water. The radial hydraulic conductivity, LPr, measures the capacity of the roots to transport water from the epidermis to the stele and can be


measured using a root pressure probe (Steudle et al. 1987). For young nodal roots of ‘Williams’, grown in solution culture, the LPr is 1.8 x 10-6 m s-1 MPa-1, which is three times higher than for the primary roots of maize under the same conditions. The LPr of maize (Gibbs et al. 1998) and banana (Aguilar et al. 2003) roots, determined hydros-tatically, decreases temporarily in response to oxygen deficiency. Gibbs et al. (1998) argue that therefore a significant flow occurs through the protoplastic pathway, even under hydrostatic gradients. This means, the measured LPr is an integrated value of all pathways of water flow. Within the protoplastic pathway, at the level of the membranes, the activity of water channels (aquaporins) is likely to change LPr (Zhang and Tyerman 1999).

Laboratory measurements, such as the above, allow maximum control. However, the question as to what happens in the field comes to mind. Sap flow technology, adapted for use in banana by Lu et al. (2002), could evaluate the hydraulic conductivity of the whole root system in the field. Miniature sap flow sensors used on fine roots of trees in field studies (Coners and Leuschner 2002) may be suited to roots of bananas and plantains.

Root signalsBanana plants have a reputation for being very sensitive to soil water deficit and there are numerous field experiments, summarized by Robinson (1996), that support this view. Banana leaves remain highly hydrated, even under drought (Shmueli 1953, Turner and Thomas 1998) and so the closure of stomata caused by soil water deficits is likely to be linked to a signal from the roots rather than water deficits in leaves. Thomas (1995), in a split root experiment, observed that drying part of the root system had no effect on leaf water status but closed stomata. Severing the roots on the dry side caused the stomata to re-open. These observations support the view that the roots produce a signal that is transported to the leaves. This mechanism conserves the plant’s water, but reduces carbon assimilation and productivity. It explains why the banana is very sensitive to soil water deficit and responsive to irrigation. A challenge will be to modify the root system so that it is less sensitive to drying soil. This would allow the plant to draw more water from the soil before growth is restricted and would increase the efficiency of water use.

Studies in root:shoot signals in bananas need to consider mycorrhizae, that can improve water uptake, and allow more water to be extracted before the root generates the signal indicating that the soil is drying. In addition, when soils dry, soil strength increases. The increased mechanical impedance may influence root signals (Masle 2002). In bananas, as soil bulk density increases, across soil types, from 0.6 to 1.2 g/cm3 root density decreases exponentially (Delvaux and Guyot 1989). Thus mechanical impedance will be an issue in studies of root signals.

Embolisms, or cavitation of xylem vessels, cause malfunction of the conducting tissues reducing the supply of water to the shoot. Embolisms occur in the roots (Shane pers. comm.) where up to 70% of the conducting vessels may be affected, even in mild climatic conditions. Stopping transpiration prevents embolisms occurring and allows those present to be refilled, presumably by root pressure, which is known to be high in bananas (Davis 1961). Modelling studies of water flow in plants and soils, reviewed


by Sperry et al. (2002), show that the amount of soil water used by a plant, especially in loamy soils, can be increased if the resistance to cavitation is increased. Studies on the sensitivity to cavitation in bananas across genotypes would establish whether this approach is likely to lead to gains in water use efficiency.

ReferencesAguilar E.A., D.W. Turner & K. Sivasithamparam. 1999. Aerenchyma formation in roots of four banana (Musa spp.)

cultivars. Scientia Horticulturae 80:57-72.

Aguilar E.A., D.W. Turner & K. Sivasithamparam. 2000a. Proposed mechanisms on how Cavendish bananas are pre-disposed to Fusarium wilt during hypoxia. InfoMusa 9(2):9-13.

Aguilar E.A., D.W. Turner & K. Sivasithamparam. 2000b. Fusarium oxysporum f. sp. cubense inoculation and hypoxia alter peroxidase and phenylalanine ammonia lyase activities in nodal roots of banana cultivars (Musa sp.) differing in their susceptibility to Fusarium wilt. Australian Journal of Botany 48: 589-596.

Aguilar, E.A., D.W. Turner, D.J. Gibbs, W. Armstrong & K. Sivasithamparam. 2003. Oxygen distribution and move-ment, respiration and nutrient loading in banana roots (Musa spp. L.) subjected to aerated and oxygen depleted environments. Plant and Soil 253:91-102.

Armstrong W. 1979. Aeration in Higher Plants. Pp. 225-332 in Advances in Botanical Research Vol. 7. (H.W. Woolhouse, ed.). Academic Press, London, United Kingdom.

Beckman C.H., M.E. Mace, S. Halmos & M.W. McGahan. 1961. Physical barriers associated with resistance in Fusarium Wilt of bananas. Phytopathology 51:507-515.

Coners H. & C. Leuschner. 2002. In situ water absorption by tree fine roots measured in real time using miniature sap-flow gauges. Functional Ecology 16:696-703.

Davis T.A. 1961. High root pressure in palms. Nature 192:277-278.

Declerck S., B. Devos, B. Delvaux & C. Plenchette. 1994. Growth response of micro-propagated banana plants to VAM inoculation. Fruits 49:103-109.

Delvaux B. 1995. Soils. Pp 230-257 in Bananas and Plantains (S. Gowen, ed.). Chapman and Hall, London, United Kingdom.

Delvaux B. & Ph. Guyot. 1989. Caractérisation de l’enracinement du bananier au champ. Incidences sur les relations sol-plants dans les bananeraies intensives de la Martinique. Fruits 44:633-647.

Draye, X. 2002. Banana roots: Architecture and genetics. Pp. 261-277 in Plant Roots: The Hidden Half (3rd edition), (Y. Waisel, A. Eshel & U. Kafkafi, eds). Marcel Dekker, New York, USA.

Gibbs D.J., D.W. Turner, W. Armstrong, K. Sivasithamparam and H. Greenway. 1998. Response to oxygen deficiency in primary maize roots 2. Development of oxygen deficiency in the stele has limited short-term impact on radial hydraulic conductivity. Australian Journal of Plant Physiology 25:759-763.

Irizarry H., S. Silva & J. Vicente-Chandler. 1981. Effect of water table level on yield and root system of plantains. Journal of Agriculture of the University of Puerto Rico 64:33-36.

Knight S. 1988. The phosphorus nutrition of bananas with special emphasis on VA mycorrhizal fungi and the effect of nitrogen. Banana Newsletter 18:1.

Lu P., K-C. Woo & Z-T Liu. 2002. Estimation of whole plant transpiration of bananas using sap flow measurements. Journal of Experimental Botany 53:1771-1779.

Masle J. 2002. High soil strength: Mechanical forces at play on root morphogenesis and in root:shoot signalling. Pp. 807-819 in Plant Roots: The Hidden Half (3rd edition). (Y. Waisel, A. Eshel & U. Kafkafi, eds). Marcel Dekker, New York, USA.

Milthorpe F.L. & J. Moorby. 1979. An Introduction to Crop Physiology. Cambridge University Press, Cambridge, United Kingdom.

Nye P.H. & P.B. Tinker. 1969. The concept of a root demand coefficient. Journal of Applied Ecology 6:293-300.

Price N.S. 1995. Banana morphology - part 1: roots and rhizomes. Pp. 179-189 in Bananas and Plantains (S. Gowen, ed.). Chapman and Hall, London, United Kingdom.

Robinson J.C. 1996. Bananas and Plantains. CAB International, Wallingford, United Kingdom.


Robinson J.C. & A. Alberts. 1989. Seasonal variation in the crop water-use coefficient of banana (cultivar ‘Williams’) in the subtropics. Scientia Horticulturae 40:215-225.

Shmueli E. 1953. Irrigation studies in the Jordan Valley 1. Physiological activity of the banana in relation to soil mois-ture. Bulletin of the Research Council of Israel 3D:228-247.

Sperry J.S., V. Stiller & U.G. Hacke. 2002. Soil water uptake and water transport through root systems. Pp. 663-681 in Plant Roots: The Hidden Half (3rd edition). (Y. Waisel, A. Eshel & U. Kafkafi, eds). Marcel Dekker, New York, USA.

Steudle E., R. Oren & E-D. Schulze. 1987. Water transport in maize roots. Measurement of hydraulic conductivity, solute permeability and of reflection coefficients of excised roots using the root pressure probe. Plant Physiology 84:1220-1232.

Thomas D.S. 1995. The influence of the atmospheric environment and soil drought on the leaf gas exchange of banana (Musa spp.). PhD thesis. The University of Western Australia, Crawley, Australia.

Turner D.W. 1994. Bananas and Plantains. Pp. 37-64 in Handbook of Environmental Physiology of Fruit Crops. 2: Sub-tropical and Tropical crops. (B. Schaffer & P.C. Andersen, eds). CRC Press, Boca Raton, USA.

Turner D.W. & B. Barkus. 1981. Some factors affecting the apparent root transfer coefficient of banana plants (cv ‘Williams’). Fruits 10: 607-613.

Turner D.W. & D.S. Thomas. 1998. Measurements of plant and soil water status and their association with leaf gas exchange in banana (Musa spp.): a lacticiferous plant. Scientia Horticulturae 77:177-193.

Turner D.W. & E. Lahav. 1983. The growth of banana plants in relation to temperature. Australian Journal of Plant Physiology 10:43-53.

Turner D.W. & E. Lahav. 1985. Temperature influences nutrient absorption and uptake rates of bananas grown in con-trolled environments. Scientia Horticulturae 26:311-322.

Zhang W.H. & S.D. Tyerman. 1999. Inhibition of water channels by HgCl2 in intact wheat root cells. Plant Physiology 120:849-857

The problem of banana root deterioration and its impact on production114 Ion absorption and proton extrusion by banana roots C.A. Gauggel et al. 115B. Delvaux et al.

Ion absorption and proton extrusion by banana rootsBruno Delvaux, Gervais Rufuikiri and Joesph Dufey1

Abstract For growth and fruit production, bananas require large amounts of nutrients, which are only partly covered by soil supply. Research on the mineral nutrition has a long history. It was first devoted to the description of symptoms of plant deficiency and the determination of fertilizer rates in a range of soils (1920s-1970s). In a second stage (1960s-2000), it focused on both the role played by nutrients in banana growth and development, and the fate of some major nutrients in soil, particularly N and K. The very high nutrient demand of bananas is expected to strongly affect soil properties in its rhizosphere wherein crucial processes occur. Surprisingly, direct interactions between banana roots and the surrounding rhizosphere have been investigated only recently. At that scale, a first step in the study of soil-plant feedback concerns ion absorption by banana roots from the liquid phase.

Calcium uptake is governed by simple convection (mass flow). Magnesium uptake seems to be only convective at low water uptake rate. Potassium, NO3

-, NH4+, P and Mn uptake rates clearly depend

on active mechanisms. For most microelements (Fe, Mo, B), exclusion mechanisms can take place in roots when supply is greater than demand.

The net proton excretion by banana roots is quantitatively related with excess uptake of cations over anions. So, the large uptake of K+ and NH4

+ ions induces strong release of acidity by banana roots.. Excretion of protons by plant roots promotes the dissolution of aluminosilicate minerals present in the surrounding rhizosphere. This dissolution liberates bio-available nutrients (Ca, Mg, Fe) as well as silicon and toxic elements such as Al. Aluminium uptake by banana roots significantly decreases plant transpiration rate, hence decreases the uptake rate of Ca and particularly Mg. This means that banana roots are able to extract a toxic element from soil minerals and take it up. The total concentration of Al is well correlated with the cation exchange capacity of banana roots. The fixation of Al on root exchange sites did not occur to the detriment of Ca but to that of Mg. The Al/Mg ratio on roots is thus a better indicator of Al-toxicity than the Al/Ca ratio. The former is significantly larger in banana roots collected in acid soils than in the ones sampled in non-acid soils.

These recent findings illustrate interactions between the soil solid phase, soil solution and banana roots. They may open a large field of research aimed at understanding the relationship between banana roots and their rhizosphere environment.

Resumen - Absorpción de iones y extrusión de protones por las raíces del bananoLa planta de banano requiere de grandes cantidades de nutrientes para su crecimiento y producción de frutos, los cuales son parcialmente suministrados por el suelo. La investigación sobre nutrición mineral tiene una larga historia. En sus inicios, ésta se dedicó a la descripción de síntomas de deficiencias y a la determinación de cantidades de fertilizantes para un rango de suelos (1920-1970). En una segunda fase (1960-2000) se investigó el rol de los nutrientes en el crecimiento y desarrollo del banano y el destino de algunos macronutrientes en el suelo, particularmente N y K. Se espera que la alta demanda de nutrientes del banano afecte grandemente las propiedades del suelo y de su rizosfera y micorrizosfera, donde ocurren procesos cruciales. Es de notar, que las investigaciones sobre las interacciones directas entre las raíces del banano y el suelo de la rizosfera que las rodean

1 Unité sciences du sol, Université catholique de Louvain, Croix du Sud 2/101348 Louvain-la-Neuve Belgium. e-mail: [email protected]


son muy recientes. A esa escala, un primer paso de interés en el estudio de las respuestas suelo-planta sería la absorción de iones por las raíces del banano en la fase líquida.

La absorción de calcio es controlada por simple convección (flujo de masas). La absorción de magnesio parece ser únicamente convectiva a una baja rata de absorción de agua. La absorción del Potasio, NO3

-, NH4-, P y Mn depende claramente de mecanismos activos. Para la mayoría de

microelementos (Fe, Mo, B), los mecanismos de exclusión pueden ocurrir en las raíces cuando hay mayor oferta que demanda.

La excreción neta de protones de las raíces de banano está cuantitativamente relacionada con un exceso de absorción de cationes sobre los aniones. De esta forma, la alta absorción de iones K+ y NH4

+, induce una fuerte descarga de acidez por las raíces de banano. La excreción de protones por las raíces de la planta promueve la disolución de minerales aluminosilicados presentes en la rizosfera vecina o adyacente. Esta disolución libera nutrientes biodisponibles (Ca, Mg, Fe) así como silicón y elementos tóxicos como el Al. La absorción de aluminio por las raíces del banano disminuye significativamente el ritmo de transpiración de la planta decreciendo así el ritmo de absorción de Ca y particularmente de Mg. Esto implica que las raíces de banano son capaces de extraer un elemento tóxico de los minerales del suelos y absorberlo. La concentración total de Al está bien correlacionada con la capacidad de intercambio de cationes de las raíces de banano. La fijación de Al en los sitios de intercambio en la raíz no sucedió en detrimento del Ca sino del Mg. De esta manera, la relación Al/Mg en las raíces es un mejor indicador de la toxicidad por Al que la relación Al/Ca. Lo anterior es significativamente mayor en raíces de banano recolectadas en suelos ácidos que en las muestreadas en suelos no ácidos.

Estos últimos descubrimientos ilustran las interacciones entre la fase sólida del suelo, la solución del suelo y las raíces del banano y pueden abrir un amplio campo de investigación para comprender la relación entre las raíces del banano y el ambiente de su rizosfera.

IntroductionResearch on the mineral nutrition of bananas first involved the description of symp-toms of plant deficiency and the determination of fertilizer rates in a range of soils (1920s-1970s) (Lahav 1995). In the second stage (1960s-1990), it focused on the role played by nutrients in banana growth and development (Lahav 1995), on soil-plant relationships with respect to cationic balance (Montagut et al. 1965, Martin-Prével and Montagut 1966), and on the fate of potassium and inorganic nitrogen in soil (Godefroy and Lossois 1966, Godefroy and Dormoy 1983, Fontaine et al. 1989).

The very high nutrient demand of bananas is expected to strongly affect the soil rhizos-phere wherein crucial processes occur, for example, water and nutrient uptake, root exudation of acidity and organic substances, invasion of roots by soil borne pathogens, establishment of mycorrhizal symbiosis (Hinsinger 1998). In particular, nutrient uptake may rapidly induce the weathering of minerals close to plant roots (Hinsinger et al. 1992). Despite of this, direct interactions between banana roots and their surrounding environment have been investigated only recently.

The aim of this paper is two-fold. First, we will report on recent studies involving inte-ractions between banana roots, the surrounding solution and clay minerals. Second, we will point to the need for further studies on the banana rhizosphere to better understand soil – banana plant interactions.


Material and methodsThe first steps in the understanding of soil-banana interactions at the rhizosphere scale concern basic knowledge on root-liquid and root-liquid-solid interactions.

Young banana plantlets from tissue culture are convenient for such studies carried out in controlled conditions. We used five varieties (AAA ‘Grande naine’, ABB ‘Kayinja’, AAB ‘Agbagba’, AAB ‘Obino l’ewaï’ and AAA ‘Igitsiri’). The plantlets were supplied by Vitropic or produced by us in the INIBAP Musa Germplasm Transit Centre loca-ted at KULeuven. Experimental procedures have been described in previous papers.

Briefly, as illustrated in Figure1, plantlets (6 repetitions) were each grown during 45 days in a 2.5 L-vessel receiving a complete nutrient flow of low ionic con-centration and known composition at a continuous rate of supply (104 ml hr-1 vessel-1) (Rufyikiri et al. 2000a). Water and nutrient uptake were measured twice a week over 24h, by weighing inlet/outlet solutions and assessing their mineral concentrations, respectively. After pH measurement, the outlet solutions were filtered at 0.45 µm. The concentrations of Ca, Mg, Fe, Mn, Cu, Mo, B and Al were determined by direct current plasma (DCP) spectrometry. Potassium concen-tration was measured by atomic absorp-tion spectrophotometry (AAS). NO3

--N and NH4

+-N were measured by Kjeldahl distillation. The net release of H+/OH- ions by roots was assessed from pH values on the basis of pH buffering curves established for inlet solutions. Plant height and leaf area were measured weekly. Dry matter was assessed after drying plant material at 60°C for 1 week. The material was ashed at 450°C for 1 day, followed by dissolution in concentrated HNO3. Elemental contents were measured as above (DCP, AAS, Kjeldahl). The cationic composition of root exchange sites was assessed (Rufyikiri et al. 2002) by extracting cations with CuSO4 (Dufey and Braun, 1986) and measuring them as above (DCP, AAS). Data obtained in hydroponics were compared with field data from bananas grown in a nutrient depleted Acrisol (pHwater = 4.2) and in a base-saturated Andosol (pHwater = 6.0). The cation exchange capacity of banana roots was measured on root axis tips and thus involved young root tissues (Dufey and Braun 1986).

Finally, root-liquid-solid interactions were studied in an experimental rhizosphere (Hinsinger et al. 1992) involving an intimate contact between roots of ‘Grande naine’ plantlets and clay minerals incorporated in agarose gel, namely Georgia kaolinite and Wyoming smectite (Rufyikiri et al. 2004). The crystal formulae of these clay minerals (previously saturated by Sr2+ cations) were:

kaolinite: {Si2Al2O5(OH)4}Sr0.001

smectite: {[Si3.96Al0.04] [Al1.52Fe0.19IIIMg0.27] O10(OH)2} Ca0.01K0.01Sr0.14.

Inlet

O2

Outlet

peristatic pumpbanana plantlet

2.5 L Vessel

Figure 1. Sketch out of the experimental design involving the continuous nutrient supply (adapted from Rufyikiri et al. 2000a).


Results and discussionThe results presented here are compiled from previous studies carried out by our group (Rufyikiri et al. 2000a, b, 2001, 2002, 2004).

Nutrient uptakeCalcium uptake by banana plantlets was governed by simple convection (mass flow) (Figure 2). Magnesium uptake seemed to be only convective at low water uptake rate, whereas active transport processes took place at high water uptake (not illustrated). Potassium, NO3

-, NH4+, P and Mn uptake rates clearly depended on active mecha-

nisms (illustrated for K, Figure 3). For most microelements (Fe, Mo, B), exclusion mechanisms could have taken place in roots when supply was greater than demand (not illustrated). The increase of ammonium uptake by banana roots resulted in a significant decrease of magnesium and the development of Mg deficiency symptoms (Rufyikiri et al. 2000a). Ammonium was selectively absorbed against nitrate, as NO3

- was weakly taken up in the presence of NH4

+ ions. Yet, upon NH4+-depletion in the nutrient solu-

tion, NO3- uptake by banana roots significantly increased (not illustrated).

10

20

30

40

50

60

0.0 0.2 0.4 0.6 0.8 1.0 1.2Water uptake (L/d)

Caup

take

(mg/

d)

Mass flow

Water uptake (L/d)

Kup

take

(mg/

d)

0.0 0.2 0.4 0.6 0.8 1.0 1.2

20

40

60

80

100

Figure 2. Calcium uptake as plotted against water uptake (convective Ca uptake as predicted from water uptake and Ca concentration in the inlet solution) (adapted from Rufyikiri et al. 2001).

Figure 3. Potassium uptake as plotted against water uptake (convective K uptake as predicted from water uptake and K concentration in the inlet solution) (adap-ted from Rufyikiri et al. 2001).

Proton excretionThe net proton excretion by banana roots was quantitatively related to excess uptake of cations over anions (Figure 4). So, the large uptake of K+ and NH4

+ ions induced a strong release of acidity by banana roots. In particular, the increase of the fraction [NH4

+/(NH4+ + NO3

-)] from 0 to 0.30 in the inlet solution at constant N supply decreased pH of from 6.5 to 3.8 in the outlet solution (not illustrated). The NH4

+/NO3-

balance thus directly influenced charge balance and proton excretion by banana roots, as it is well known for other plant species (Marschner 1995). However, banana plant-lets were remarkable in the amplitude of H+-excretion as assessed by pH measurement in the outlet solution, since the above-mentioned increase of NH4

+/NO3- involved an

increase of proton activity of three orders of magnitude (pH 6.5-3.8) in the outlet solu-tion. This was likely due to both the large uptake of cations (K+, NH4

+) and the rapid growth of banana plantlets in controlled conditions (Rufyikiri et al. 2001).


Mineral dissolution around rootsExcretion of protons by plant roots promotes the dissolution of aluminosilicate minerals present in the surrounding rhizosphere (Hinsinger 1998). This dissolution liberates bio-available nutrients (Ca, Mg, Fe) as well as silicon (Hinsinger et al. 2001).

As reported by Rufyikiri et al. (2004), protons excreted by banana roots were neu-tralised by the dissolution of clay minerals such as kaolinite and smectite. However, H+ neutralization was much more effective through smectite dissolution because this mineral was less stable than kaolinite in the experimental conditions that prevailed. The root induced dissolution of clays released Al from kaolinite and smectite, but also released Mg from smectite. These elements were readily taken up by banana roots over time (Figure 5). This means that banana roots were able to extract toxic Al from clays and absorb it.

-200

-100

0

100

200

300

400

500

-200 -100 0 100 200 300 400 500Balance of anion-cation uptake (�mol/d)

Net

root

release(�mol/d

)OH-H+

y = -16.96 + 1.04 x (r=0.99)

Figure 4. Relationship between net H+/OH- release by banana roots and cation-anion balance of uptake (adapted from Rufyikiri et al. 2001).

0.8

1.2

1.6

2.0

0 1 2 3 4 5 6

Conten

ts(g/k

gdryweigh

t)

Magnesium

Time (weeks) Time (weeks)

0.0

0.10

0.20

0 1 2 3 4 5 6

Aluminium

controlkaolinitesmectite

Figure 5. Week variation of the Mg and Al contents of banana roots growing in the close vicinity of smectite and kaolinite clay minerals (adapted from Rufyikiri et al. 2004).


Ion adsorption on root exchange sitesFor the five banana varieties investigated, the cation exchange capacity of roots (CECR) differed between root axes (23 ± 4 cmol/kg root) and lateral roots (34 ± 3 cmol/kg root) (Rufyikiri et al. 2002). This was likely to be related to the lower contribution of secondary walls to dry matter in small lateral roots.

Aluminium was selectively adsorbed on banana root exchange sites against divalent cations. However, Al fixation on root exchange sites did not occur to the detriment of Ca, but to the detriment of Mg (Rufyikiri et al. 2002, Rufyikiri et al. 2003). The Mg propor-tion dropped from 40% to 5% when Al was competing for ion exchange on root sorbing sites, whereas the Ca proportion remained at about 60% (Figure 6). Similarly, the pro-portion of Al on root exchange sites was larger for banana roots sampled in the nutrient-depleted Acrisol than in the base-saturated Andosol. These observations may explain why Mg-deficiency symptoms associated with the ‘bleu magnésien’ occur preferentially in acid soils (Montagut et al. 1965; Martin-Prével and Montagut 1966).

In the base-saturated Andosol, the sum of Cu-extractable cations (Al, Ca, Mg) amounted up to 32 cmolc/kg root because of higher pH (6.0 against 4.2 in the Acrisol). Despite this pH value, Al was readily sorbed on exchange sites of roots (Figure 6), and thus likely mobilised from mineral dissolu-tion in the vicinity of plant roots. Moreover, this also indicates that Al is very efficient in compe-ting with other cations for root exchange sites despite the low Al concentrations usually found in soil solutions (Dufey et al. 2001).

Al uptake and toxicityAluminium uptake by banana roots significantly decreased plant transpiration rate, hence decreased the uptake rate of Ca and particularly Mg, which are transferred to plants chie-fly by mass flow. The toxicity of Al to bananas significantly reduced dry matter, stunted the roots and induced Mg deficiency (Rufyikiri et al. 2000a, Rufyikiri et al. 2001).

Arbuscular mycorrhizal fungi (AMF) might alleviate Al toxicity to bananas. The in vitro symbiotic association of AMF (Glomus intraradices) and micropropagated bana-na plants of ‘Grande naine’ indeed increased plant growth and dry matter, water and nutrient uptake, whereas it involved a decrease in Al concentration of plants as well as a delay in appearance of Al-induced leaf symptoms (Rufyikiri et al. 2001).

ConclusionsThe large nutrient demand of banana represents a very strong potential for soil net acidification, which is directly governed by the balance of cation-anion uptake. In par-

0

10

20

30

40

-Al + Al Andisol Acrisol

Cu-extractab

leca

tion

scm

ol/k

groot

dryweigh

t Ca Mg Al

Figure 6. Cu-extractable contents of Al, Mg, Ca of banana root exchange sites from bananas cultivated in controlled conditions (con-tinuous flow device: Figure 1) as well as in field conditions (Andosol, Acrisol) (adapted from Rufyikiri et al. 2002, 2003).

120

ticular, the large uptake of potassium and the selective absorption of ammonium against nitrate play a major role in acid excretion by banana roots.

Protons excreted by banana roots dissolve secondary soil minerals that commonly occur in soils (kaolinite, smectite). Clay minerals as stable as kaolinite can be dissol-ved over short time periods (weeks). Proton excretion exceeds the acid neutralizing capacity of the clay minerals close to banana roots. Mineral dissolution liberates Al and Mg previously occluded in crystal structures. Aluminium uptake by banana roots reduces water uptake, nutrient uptake and particularly Mg uptake. Selective Al fixation on banana root exchange sites has a strong detrimental effect on Mg sorption, and hence uptake. Aluminium toxicity to bananas might be alleviated by symbioses involving arbuscular mycorrhizal fungi.

PerspectivesThe recent findings presented in this paper illustrate a few of the interactions between soil solid phase, soil solution and banana roots. They may open a large field of research aimed at understanding the relationship between banana roots and their rhizosphere.

With respect to the scope of this symposium, this paper modestly contributes to the need for pursuing deeper investigations on the rhizosphere of banana plants. As shown above, banana roots may exert a strong influence on their surrounding environment. Intensive banana cropping systems, particularly in prolonged monocultures, may thus exhaust soils on a presently little-known scale.

The rhizosphere environment is very unique in terms of physicochemical properties and processes. Interactions between these properties/processes and biotic processes and constraints are poorly known in soils long influenced by intensive banana cropping. Unpublished data show that both the pathogenic nematode Radopholus similis and fungus Cylindrocladium spp. penetrate the root between the apex and the first lateral root as well as through the locations of extrusion of lateral roots. R. similis preferenti-ally invade banana roots excreting alkalinity, i.e. taking up large amounts of NO3

- over NH4

+ (Declerck et al. 1998). Cylindrocladium spp. preferentially invade roots of Mn-deficient banana plants (unpublished data). In this respect, the continuous solution flow device allows us to study root-liquid-solid interactions in controlled conditions. This system is convenient for studying symbiotic interactions at the scale of the ‘experi-mental rhizosphere’ in controlled conditions of plant nutrition. It is also promising for the study of root-pathogen interactions in controlled conditions. Yet, in vitro studies are best coupled with field investigations. In this respect, we need a reproducible and standard procedure to accurately sample the banana rhizosphere

ReferencesDeclerck S., S. Laloux, J.L. Sarah & B. Delvaux. 1998. Application of a flowing solution culture technique to study

the parasitic fitness of the nematode Radopholus similis on banana plantlets under two different nitrogen nutrient regimes. Plant Pathology 47:580-585.

Dufey J.E. & R. Braun. 1986. Cation exchange capacity of roots: titration, sum of exchangeable cations, copper adsorp-tion. Journal of Plant Nutrition 9:1147-1155.

Ion absorption and proton extrusion by banana roots


Dufey J.E., J.G. Genon, B. Jaillard, H. Calba, G. Rufyikiri & B. Delvaux . 2001. Cation exchange on plant roots involving aluminium: experimental and modelling. Pp. 228-252 in Fate of trace elements in the rhizosphere (G.R. Gobran, W.W. Wenzel & E. Lombi, eds). CRC Press LLC, Boca Raton, Florida.

Fontaine S., B. Delvaux, J.E. Dufey & A.J. Herbillon. 1989. Potassium exchange behaviour in Caribbean volcanic ash soils under banana cultivation. Plant and Soil 120:283-290.

Godefroy J. & M. Dormoy. 1983. Dynamique des éléments minéraux fertilisants dans les sols des bananeraies martini-quaises. Deuxième partie. Fruits 38:451-459.

Godefroy J. & P. Lossois. 1966. Variations saisonnières des caractéristiques physico-chimiques d’un sol volcanique du Cameroun. Fruits 21:535-542.

Hinsinger P. 1998. How do plant roots acquire mineral nutrients? Chemical processes involved in the rhizosphere. Advances in Agronomy 64: 225.

Hinsinger P., B. Jaillard & J.E. Dufey. 1992. Rapid weathering of a trioctahedral mica by the roots of Rye-grass. Soil Science Society of America Journal 56:977-982.

Hinsinger P., O.N. Fernandes Barros, M.F. Benedetti, Y. Noack & G. Callot. 2001. Plant-induced weathering of a basal-tic rock: experimental evidence. Geochim. Cosmochim. Acta 65:137.

Lahav E. 1995. Banana nutrition. Pp. 258-316 in Bananas and Plantains (S. Gowen ed.). Chapman & Hall, London, UK.

Marschner H. 1995. Mineral Nutrition of Higher Plants (2nd edition). Academic Press, London, UK.

Martin-Prével P. & G. Montagut. 1966. Essai sol-plante: les interactions dans la nutrition minérale du bananier. Fruits 21:19-36.

Montagut G., P. Martin-Prével & J.J. Lacoeuilhe. 1965. Essais sol-plante. Nutrition minérale comparée dans six essais. Fruits 20:398-410.

Rufyikiri G., J. Dufey, D. Nootens & B. Delvaux. 2000a. Effect of aluminium on bananas (Musa spp.) cultivated in acid solutions. I . Plant growth and chemical composition. Fruits 55:367-379.

Rufyikiri G., S. Declerck, J.E. Dufey & B. Delvaux. 2000b. Arbuscular mycorrhizal fungi might alleviate aluminium toxicity in banana plants. New Phytologist 148:343-352.

Rufyikiri G., D. Nootens, J.E. Dufey & B. Delvaux. 2001. Effect of aluminium on bananas (Musa spp.) cultivated in acid solutions. II. Water and nutrient uptake. Fruits 56:3-14.

Rufyikiri G., J.E. Dufey, R. Achard & B. Delvaux. 2002. Cation exchange capacity and Al-Ca-Mg binding in roots of bananas (Musa spp.) cultivated in soils and in nutrient solutions. Communications in Soil and Plant Analysis 33:991-1009.

Rufyikiri G., J.G. Genon, J.E. Dufey & B. Delvaux. 2003. Competitive absorption of H, Ca, K, Mg and Al on banana roots: experimental data and modelling. Journal of Plant Nutrition 26:351-368.

Rufyikiri G., D. Nootens, J.E. Dufey & B. Delvaux. 2004. Mobilization of aluminium and magnesium by roots of banana (Musa spp.) from kaolinite and smectite clay minerals. Applied Geochemistry 19:633-643.

B. Delvaux et al.


4

Soils and root development

Suelo y desarrollo radical

The problem of banana root deterioration and its impact on production124 Soil physical properties and banana root growth C.A. Gauggel et al. 125R. Vaquero

Soil physical properties and banana root growthRoque Vaquero M1.

AbstractThe physical properties of the soil regulate the conditions in which banana roots grow; therefore, they should be among the criteria evaluated to determine the soil’s potential for banana production. Under adequate nutritional conditions and with a good water supply (rain or irrigation), banana roots require aeration and low mechanical resistance for normal growth. Therefore, the physical characteristics that affect the water-air balance and soil consistence (the resistance of a soil to mechanical stress at various water contents), are manifested in the growth of the plant root system and consequently in crop productivity.

Resumen - Propiedades físicas del suelo y desarrollo del sistema radicalLas propiedades físicas del suelo regulan las condiciones del medio en que crecen las raíces del banano, por lo que se deben incluir dentro de los criterios para valorar el potencial del suelo para el cultivo. Bajo condiciones nutricionales adecuadas y con un suministro apropiado de agua (lluvia o riego), las raíces del banano necesitan aireación y baja resistencia mecánica, que les permitan un crecimiento normal. Por lo tanto, el efecto de las características físicas que afectan el balance agua-aire y la consistencia del suelo (la resistencia del suelo al stress mecánico en varias condiciones de humedad), se refleja en el crecimiento del sistema de raíces de la planta y por consiguiente en la productividad del cultivo.

Introduction Texture, compaction and drainage are the main physical soil characteristics that influence banana growth and development; they can limit effective soil depth or the conditions of water supply and aeration in the rhizosphere. According to Stover and Simmonds (1987), studies carried out in bananas indicate that root distribution, in relation to soil depth, is determined by soil type and drainage conditions and that root growth is stopped or reduced by hardpans, impermeable layers, high clay content or waterlogged areas.

Layers or horizons within the coarse textural class (sandy soil) and the very fine (clayey, with more than 60% clay) hinder the development of the banana root system, due to their effect on water retention capacity, permeability and the water-air balance. The presence of coarse fragments (above 15% by volume) is also considered a limi-ting factor in banana root growth. Compact soil layers limit root growth and affect the properties related to water and air movement. Soil structure and consistence are related to compaction and an adequate field evaluation of them facilitates the diagnosis of the physical conditions of the root environment.

Horizons in the soil profile that indicate high water saturation, either permanent or temporary, are indicators of conditions that restrict root growth and can cause necrosis and rotting during very wet spells (Lara 1970 cited by Soto 1990, Dorel 1993, Stover and Simmonds 1987).

1Roque Vaquero Morris. Professor at EARTH University, Costa Rica. e-mail: [email protected]


Soil texture and root growth Commercial banana farms are located in humid climates with a relatively flat topo-graphy. The most extensive farms are on alluvial plains, with some exceptions in dry climates. The rest are located in small areas in soils of volcanic origin or of varied mineralogy, on irregular topography, others on peat over clay. In other areas bananas are planted on bench terraces with transported soil (Stover and Simmonds 1987).

In the American tropics, in alluvial areas, soil texture has been one of the principal factors used to select the soil for the crop (Stover and Simmonds 1987). The spatial variability of soil texture in alluvial lands is determined by the gradient in particle size that varies transversally and longitudinally according to the location of the stream flow where it originated. Close to the stream and on higher grounds, the soil is of coarse texture or with higher sand content and lower clay content than those that are found downstream or further from the riverbank. Therefore, the best areas for the crop on allu-vial plains are in medium-textured soils near streambeds. Soils with high sand or clay contents have low potential or are inadequate for bananas. The former due to their low water retention capacity and natural fertility and the latter due to their low permeability and because they are generally found in low areas, that consequently, present drainage problems.

However, there are banana plantations in clayey soils and low-lying areas that have excellent growth and reach high production. It is evident that the dominating effects of clay-sized particles, the fine fraction, can be outweighed by a moderate to strong structure and a high potential fertility due to their material of origin deposited on some river floodplains (as observed in Costa Rica, Panama, and in some areas of Guatemala). According to Soto (1990), the best soil textures for obtaining high banana production go from very fine and fine sandy loam to clay loam, meanwhile finer or coarser textures lead to management problems. Soto also points out that light textures are favourable in the subsoil to assist with drainage, and that plants in medium to finer textured soils with good structure and porosity, develop a more extensive and deeper root system.

At the commercial banana plantation at EARTH University, in the Atlantic region of Costa Rica, the author made soil descriptions and root mapping, in profiles 1.5 m wide and 1.0 m deep, in semicircles 50 cm from the base of the banana plant, in front of the sucker and at the time of shooting, selecting the sites based on the soil classification made by Sanchez (1994). One of these tests was on a class IV soil with 17% of the roots located below 50 cm, while better root distribution was noticed in a plant from the same clone planted on a class II soil, with more than 30% of the roots below 50 cm. The plant surpassed, by more than 15%, the plant in class IV soil, in height and circumference of the pseudostem at shooting (Table 1).

There was a higher root density in loam to silty loam soils, decreasing as the sand content increased to textures like sandy loam (Figure 1). Lower root densities were observed in soils with higher clay content (clay loam) and a finer texture (silty clay).


Soil compaction and root growth limitationsSoil compaction results when the mass, upon compression, occupies part of the volume of the air spaces. Thus, one of the major effects of compaction is manifested in decreased soil permeability and increased physical resistance.

Most of the research work that relates soil physical properties to banana root growth has been done in Guadeloupe and Martinique. The studies carried out by Gousseland (1983) established the relation between bulk density and root growth under controlled conditions, finding a direct relationship between them. Work done by Delvaux and Guyot (1989), in different soil types with bulk density between 0.7 and 1.1 g/cm3 indicates that when the bulk density of a soil is above 0.95 g/cm3, root density is significantly reduced. The author has obtained data on bulk densities above 1.15 g/cm3,in fine texture soils, and 1.35 g/cm3 or more in medium to medium-coarse textured alluvial soils in the Sula Valley, Honduras, derived from sediments of varied mineralogy. And there, banana plantations do not show the effects of soil compaction problems.

4,14,6

5,1

3,9

2,5

0,00

1,00

2,00

3,00

4,00

5,00

6,00

L-TL L-SL CL CL-C TC

Loam-Silty loam Loam-Sandy loam Clay loam Clay loam-Clay Silty clay

Soil texture

Root

density(roo

ts/d

m2

Figure 1. Average root density on vertical cut of the first 25 cm of soil depth, with different textures, sampled in front of a sucker of a flowering banana plant. EARTH commercial farm, 2003.

Table 1. Vertical root distribution, in front of banana plants of the ‘Williams’ clone, at flowering, in two soils with different banana production potential. EARTH commercial Farm, September 2003. Depth (cm) Soil Class II Soil Class IV

Roots Texture* Roots Texture*

Quantity % Quantity %

0-25 173 42 L 148 45 CL25-50 114 28 L 124 38 C50-75 72 17 SL 31 9 C75-100 55 13 LS 25 8 CPlant height (cm) 345 299Pseudostem circumference (cm) 74 65* L = loam; SL = fine sandy loam; LS = fine loamy sand; CL = clay loam; C = clay.


Dorel (1993) argues that although it is difficult to evaluate the effect of soil compaction on banana production, soil compaction reduces plant and bunch size, delays flowering and reduces sucker emergence. Experiments carried out in andosols in Martinique demonstrate that compaction reduces aeration and leads to root death due to lack of oxygen and its immediate effect is reflected in the decline in the number of roots and, consequently, nutritional imbalances in the plant, because of a reduced capacity of the root system to absorb nutrients.

Mechanical penetration resistance of the soil is a result of the consistence and com-paction of the soil horizons. Some empirical values have been established for samples taken with a pocket penetrometer, Soiltest CL-700A. When the soil is at field capacity, penetration resistance values of 2.0 kg/cm2 or lower, are associated with a friable to very friable soil consistence, which is adequate for banana root growth. Values higher than 2.0 kg/cm2 and close to 3.0 kg/cm2 are associated with a firm soil consistence and indicate a higher soil compaction and a certain degree of root growth limitation. Meanwhile, values close to, or higher than 3.5 kg/cm2 have been obtained for firm to very firm soil consistence that severely limits banana root growth.

The author’s experience in different soil types in Honduras and Belize has shown that penetration resistance can be utilized as a parameter to evaluate soil compaction, taking into consideration that the degree of compaction is affected by other variables, like soil texture, structure and water content. The conversion of the penetrometer reading to rela-tive values is also adequate to explain the difference in banana root growth in different sites and soil profiles. There is a relationship between soil consistence and the pene-tration resistance values measured with a penetrometer (Soiltest CL-700A) (Table 2). A range of soil consistence and a range of soil penetration resistance and bulk density at each depth in the soil profile can identify the principal cause of the lower values of root density (Table 3). Root density of banana plants is higher in friable to very friable soils with low bulk density and penetration resistance values (<2.5 kg/cm2) (Table 3). But, root density is drastically reduced in areas with high penetration resistance values (>2.5 kg/cm2) associated with a firm or very firm soil consistence and bulk density above 1.2 g/cm3. In alluvial soils the author has used a different approach to evaluate the problem of compaction by considering the relation between structure development and soil consistence (Table 4).

Table 2. Penetration resistance (kg/cm2)* for different soil texture and consistence condi-tions in vertical profiles utilized for root description. EARTH commercial farm, 2003. Soil texture Soil consistence

Very friable Friable Firm Very firm

Coarse 0.9 – 1.5 ** 4.3 - > 4.5 **

Medium coarse 1.8 – 1.3 ** 2.8 – 3.5 > 4.5

Medium 0.3 – 1.3 0.5 – 1.5 ** **

Medium fine 0.8 – 1.5 0.6 – 2.5 ** **

Fine ** 0.8 – 2.5 2.3 – 3.5 *** Reading ranges from a pocket penetrometer (Soiltest CL-700A) down to 25 cm depth, with readings at intervals of

5 cm. ** These conditions were not present.


Soil drainage and root growthSaturated soil and shallow water tables reduce the number of roots, root system growth and banana plant productivity (Ghavami 1976, Holder and Gumbs 1983, cited by Stover and Simmonds 1987). In the American tropics, bananas are grown in regions with medium to high precipitation leading to temporary or permanent problems of

Table 3. Root density (Rd roots/dm2), soil consistence C, penetration resistance (Rp kg/cm2) and soil bulk density (Db g/cm3) in soil layers, each with a 25 cm depth, for EARTH commercial farm, 2003. Depth Rd C* Rp ** Db*** Minimum value of(cm) root density related to:

0-25 2.5 - 5.1 vFr - Fr 0.4 – 2.5 0.72 – 1.17 Fine texture

25-50 0.9 – 3.3 vFr - Fr 0.4 – 1.8 0.62 – 1.12 Indicates poor drainage

50-75 0.5 - 2.5 vFr - vFir 0.25 - >4.5 0.93 – 1.11 Compaction and poor drainage in medium texture; fine texture

75-100 0 - 1.7 vFr - Fir 0.91 - >4.5 0.91 – 1.24 Poor drainage, consistency and compaction* vFr = very friable; Fr = friable; Fir = firm; vFir = very firm. ** Penetration resistance; reading range with a pocket penetrometer (Soiltest CL-700 A). *** Bulk density of the soil.

Table 4. Scores for interpreting the interaction between soil consistence and structure. d.a. = does not applySoil Structure development grade

consistence Strong Moderate Weak Massive

Loose d.a. d.a. d.a. d.a.

Very friable 1 1 1 2

Friable 1 1 1 2

Firm 2 2 3 4

Very firm 3 3 4 4

Extremely firm 3 3 4 4

Interpretation of the soil consistence x structure scores

Score Interpretation General actions

1 Low or unnoticed No corrective measures are expected, but measures should be undertaken to maintain these conditions

2 Moderate Requires some soil/water management practices (mechanized preparation, organic matter incorporation, drainage, use of cover crops and green manure, amongst others)

3 Severe Requires specific soil/water management practices (plowing and sub soiling, incorporating high amounts of organic matter, alternate crop cycle with fallow and cover crops to break compact layers, incorporate green manure, intensive drainage practices, special irrigation control, and others)

4 Very severe Even if intensive management practices are put in place, conditions might not improve, therefore species, cultivation system, physical inputs and agricultural techniques must be precisely selected


waterlogging in the soil profile. In these areas, excessive runoff is produced and the water table rises to a depth that directly affects plant root depth and density.

Some research indicates that water tables in banana farms should not be shallower than 120 cm. They also show that deeper water tables are associated with a better perfor-mance of production variables (Ghavami 1976, Sancho 1991), because the air-water balance is improved, which facilitates root growth. Stover and Simmonds (1987) argue that banana crops require good surface and internal drainage and that the watertable should preferably be below 1.2 m deep. They also report that watertable fluctuations cause damage when the water rises to the dense rhizosphere and remains there for more than 24 hours. According to Stover and Simmonds (1987) root horizon profiles are good indicators of poor drainage or compact subsoil, and low permeability and that, in Central America, it is clear that there are factors that limit root growth (mainly drainage) when the percentage of roots below 60 cm is less than 25% of the plant root system.

In recent root readings done by the author and mentioned before, a lower density was found in soils with drainage problems and in horizons with evidence of temporary water saturation, due to low permeability caused by compaction (Table 5). Water satu-ration on a permanent (hydromorphic horizons) or variable (mottled horizons of low permeability) basis represent a huge limitation for root growth (Table 5). Soil profiles that do not reveal signs of poor drainage in the first meter of depth have a better root distribution with depth. Additionally, because the majority of the banana root system is located in the first 30 - 50 cm and, due to its high susceptibility to water excess or deficit, it is important to avoid ponding and waterlogging that can lead to root necrosis by lack of oxygen or toxicity by some chemical compounds that are more soluble under these conditions.

Table 5. Root density in soil profiles with different drainage conditions. EARTH commercial farm, 2003. Soil profile number and drainage

Depth (cm) 2, poor 5, good 3, good 4, moderate

Root density, root/dm2

0-25 2.5 4.1 4.6 5.1

25-50 0.9 2.8 3.0 2.6

50-75 0.5 2.5 1.9 1.7

75-100 0.0 1.7 1.5 0.0Profile 2: class III, poor drainage, water table at 1.10 m, hydromorphic horizons below 75 cm; clayey on the surface and clay loam in the subsoil; grey and red mottling along the profile; firm layer from 50 – 75 cm in depth.Profile 3: Class II, well drained, deep; loam to fine sandy loam; friable to very friable; higher sand content below 50 cm.Profile 4: Class III, moderate drainage with evidence of temporary water saturation below 50 cm and a firm layer with low permeability at a depth of 75 – 100 cm. Profile 5: Class II, well drained up to 1.10 m, below hydromorphic layer; clay loam, friable to very friable, with a firm layer (5 cm thick) below 85 cm depth.


Final remarksSoil texture, the degree of compaction and the drainage conditions affect banana root growth and density. The effects of these variables cannot be addressed on an individual basis because of the multiple interactions they have on the root stratum. The efforts that are undertaken to analyze the soil’s physical characteristics in order to evaluate its potential for this crop, are intended to help locate areas that can guarantee high pro-duction with a reasonable investment, minimizing the risks of this activity. Research and the development of techniques to evaluate and solve soil compaction problems, improve structural conditions, soil stratum consistence and porosity that allow the plant root system to develop are ways to introduce improvements to increase productivity in cultivated areas.

Finally, the development of strategies and schemes to avoid surface ponding through soil conformation or surface drainage, and proper design of major drainage infrastruc-ture to prevent watertables from rising to the root zone, can help guarantee a healthy and deeply-developed root system and, as a consequence, high production.

ReferencesDelvaux B. & Ph. Guyot. 1989. Caractérisation del enracinement de bananier au champ. Incidences sur les relations

sol-plante dans les bananeraies intensives de la Martinique. Fruits 44:633-647.

Dorel M. 1993. Développement du bananier dans un andosol de Guadeloupe: effet de la compacité du sol. Fruits 48:83-88.

Ghavami M. 1976. Banana plant response to water table levels. Transactions of the American Society of Agricultural Engineers 19:675-677.

Gousseland J. 1983. Etude de l’enracinement et de l’émission racinaire du bananier ‘Giant Cavendish’ (Musa acuminata AAA sous-groupe Cavendish) dans les andosols de la Guadeloupe. Fruits 38:611-613.

Sanchez L. 1994. Estudio de reclasificación de suelos de la finca EARTH y programa de fertilización.46pp.

Sancho H. 1991. Respuesta del banano (Clon Valery) a tres condiciones de drenaje. Pp. 41-50 in Memorias ACORBAT X (M.A. Contreras, J.A. Guzmán & L.R. Carrasco, eds). Villahermosa, México.

Soto M. 1990. Bananos: Cultivo y Comercialización. 2nd ed. San José, Costa Rica.

Stover R.H. & N.W. Simmonds. 1987. Bananas. 3rd ed. Tropical Agriculture Series. John Wiley & Sons, Inc. New York, USA.

The problem of banana root deterioration and its impact on production132 Interrelations between the soil chemical properties and the banana plant root system C.A. Gauggel et al. 133C.A. Gauggel et al.

Interrelations between the soil chemical properties and the banana plant root systemCarlos A. Gauggel1, Diana Moran1 and Eduardo Gurdian1

AbstractThe relationship between the solid phases, organic and non-organic, liquid (soil solution) and the banana plant root system have been found to influence soil chemical conditions that determine commercial yields. The role that the minerals, found in clay soils that are planted with bananas in Latin America, have on the activity of the elements and substances in soil solution and their availability to the banana plant is examined. Some of the most important chemical properties are: the availability of phosphorus in the allophanic soils of the Pacific coast of Central America; saline, sodic and saline-sodic soil conditions of the Santa Marta, Colombia zone; and the highly available iron, manganese and exchangeable aluminum concentrations in the soils along the Atlantic coasts of Costa Rica and Urabá, Colombia. The importance of improving knowledge about the banana plant rhizosphere as a key factor to better root health management and improve yields is highlighted. Technologies such as mycorrhizae (VM) offer opportunities to substantially improve banana root systems in sub-optimal soil conditions.

Resumen - Interrelaciones entre las propiedades químicas del suelo y el sistema radical del bananoLas interrelaciones entre las fases sólida, orgánica e inorgánica coloidal, liquida (solución del suelo) y el sistema radical del banano son conocidas por inferir en las condiciones químicas del suelo que afectan el rendimiento comercial del cultivo. Se examina el papel que juega la mineralogía de suelos arcillosos sembrados con banano en América Latina en la actividad de elementos y sustancias de la solución del suelo y su disponibilidad a la planta del banano. Algunas de las propiedades químicas mas importantes son: la disponibilidad del fósforo en los suelos con contenidos altos de alófano en las costas del Pacifico Centroamericano; las condiciones que presentan los suelos sódicos, salinos y salino sódicos en la zona de Santa Marta, Colombia; y la alta disponibilidad del hierro, manganeso y el contenido de aluminio intercambiable en los suelos de la Costa Atlántica de Costa Rica y Urabá, Colombia. Se destaca la importancia de incrementar el conocimiento de los diferentes componentes de la rizosfera del banano como un factor para mejorar el manejo de la salud del sistema radical y la producción. Las tecnologías como el uso de las micorrizas (MVA) ofrecen oportunidades para mejorar sustancialmente el desempeño del sistema radical del banano en condiciones de suelos deficientes.

IntroductionThe relationship between soil chemical properties and the banana root system has great practical importance. Unfortunately, this relationship has not been fully understood and its economic impact has not been accurately estimated due to the complexity of its mul-tiple components. This lack of understanding brings negative economic consequences since accurate and effective soil management programs cannot be designed; thus limi-ting crop yields to different degrees, depending upon the severity of the limitation.

A fair level of understanding has been reached to predict the behavior of nitrogen, magnesium, copper, zinc, boron, sodium and soluble salts, given the relative simpli-

1 Panamerican Agricultural School, El Zamorano, Honduras. e-mail: [email protected]


city of their reactions in the soil system. However, the level of understanding of the relationship between soil reaction, exchangeable aluminum, iron, manganese, organic substances, potassium and phosphorus is more limited. Nevertheless, it is known that each of them has a high degree of complexity in the interaction between the soil (orga-nic and inorganic colloids), liquid (soil solution) and the rhizosphere of the banana and plantain (Table 1).

It is clear that a thorough understanding of these interactions will lead to more accurate and effective programs that will substantially increase yields. Actually, due to the lack of specific studies in bananas and plantains, many predictions and recommendations are based on more general knowledge of soil-plant relationships.

Table 1. Level of understanding of the relationship between the chemical soil properties and the root system of the banana plant.Satisfactory level Partial level Complex factors necessaryof understanding of knowledge to understand the soil-root system relationship

pH Complexity of the causes of acidity or different sources of alkalinity

Kinds and contents Lack of good and cheap methods to characterize organic compounds of organic matter and their physic-chemical processes

Potassium Behavior and availability of potassium with different types of clay

Phosphorus Complexity of phosphorus reactions with the different primary minerals of the soil, iron and magnesium oxides and hydroxides, changes in pH and clay contents.

Aluminum Susceptibility of the different banana varieties to the possible toxicity of aluminum and understanding of the processes at the cellular level in the banana plant

Iron and Manganese Oxidation states of iron and magnesium in the soil and the level of susceptibility of the different banana varieties

Cation Exchange Capacity pH dependent charges, different composition of the solid fraction of the soil phase (clay and organic matter)

Nitrogen

Magnesium

Copper

Zinc

Boron

Sodium

Soluble salts

Soil phases and the banana plant root systemThe solid phaseThe inorganic component: primary minerals Soils devoted to banana production in Latin America (including Martinique and Guadeloupe) are composed of parent materials with diverse primary minerals. Those, such as the slightly weathered minerals from the active volcanic zones, metamorphic or sedimentary rocks rich in quartz, feldspar and mica weathered to various degrees, and calcium carbonate from areas rich in limestone, have the greater influence on the rela-tionship between soil chemical properties and the banana plant rhizosphere (Table 2).

The release of sodium from albite (sodium feldspar) bears special importance in the soils of the banana zones of Santa Marta, Colombia, the alluvial soils of the Yaque del Norte River in the Dominican Republic and south of Maracaibo Lake in Venezuela. This causes high contents of exchangeable and soluble sodium in many soils that, in


turn, cause severe toxicity problems in bananas and plantain. Soils formed from volca-nic sediments (Andisols) rich in allophane are notorious for fixing phosphorus. These soils are typical of the Pacific Coasts of Central America and large parts of Martinique and Guadeloupe. Limestone soils, rich in calcium carbonate, also fix considerable amounts of phosphorus and restrict the availability of micronutrients and result in magnesium and potassium imbalances. Large areas of such soils are common in the northwest alluvial valleys of Honduras, Belize, some areas of the Dominican Republic, Santa Marta in Colombia and Martinique.

In this way, primary minerals regulate the availability of phosphorus, magnesium, potassium and most micronutrients for plant uptake. These specific cases have not been widely studied in bananas and plantains; however, the crop response to the application of specific nutrients highlights the importance of the reactions indicated above in the geographic areas of interest.

The inorganic fraction: phyllosilicates Together with humus, phyllosilicates (clay fraction of soils and sediments) determine the kind and quantity of elements entering the soil solution that eventually can be taken up by plants. High contents of smectite restrict the availability of phosphorus and potassium in the soils of the Sula Valley in Honduras (Table 3). In the case of phospho-rus, this is aggravated by the relatively high calcium carbonate content of these soils. Besides, it also alters the (Ca+Mg)/K and Ca/Mg ratios and the availability of most

Table 2. Predominant soil mineralogy of the main banana production zones of Latin America.Geographic zone Minerals of the coarse fraction Secondary minerals of the soil

Alluvial valleys of Honduras Quartz, feldspar and calcium Smectite, vermiculite, carbonate kaolinite and illite

Atlantic coast of Colombia Quartz, degraded muscovite Illite, vermiculite, smectiteSanta Marta and sodium feldspar (albite) and kaoliniteUrabá Quartz Kaolinite, vermiculite, illite and smectite

Atlantic coast of Costa Rica Quartz, volcanic amorphous Allophane, imogolite, kaolinite, materials and calcium carbonate vermiculite and sesquioxides

Atlantic coast of Guatemala Quartz, feldspar and degraded Kaolinite, vermiculite and smectite muscovite

Atlantic coast of Panama Quartz Vermiculite, kaolinite, illite and allophane

Dominican Republic Quartz, calcite and sodium Vermiculite and smectite. feldspar (albite)

Pacific coast of Guatemala Recent primary minerals Allophane

Pacific coast of Nicaragua Recent primary minerals Allophane

Martinique and Guadeloupe Recent primary minerals and Allophane and smectite calcium carbonate

Venezuela, Degraded muscovite, sodium Illite, vermiculite, smectite and Golf of Maracaibo quartz and sodium feldspar kaolinite (albite)


micronutrients. The same condition is common in some areas of the Yaque del Norte River in the Dominican Republic and Trois-Ilets in Martinique, and to a lesser extent in the Machala and Los Rios provinces in Ecuador. The very low potassium concentration in the colluvial-alluvial soils of the Sierra Nevada in Santa Marta, south of Maracaibo Lake (Venezuela), and some soils of the Machala and Los Rios provinces of Ecuador is determined by high contents of illite in the soil clay fraction (Table 3). This condition is aggravated in soils with high clay content since the higher the clay content the higher their potassium fixing potential.

High contents of allophane occur in the soils of the Pacific coast of Central America. Besides fixing phosphorus, the cation exchange capacity is highly dependent on the pH of the soil solution. Thus, when fertilizers are applied and the pH of the soil solution changes, the cation exchange capacity of the soil is altered, resulting in changes in the nutrient availability for plants. A special case occurs in the extensive alluvial system of the banana zones of Ecuador where sediments originate from various sources. In this particular case, the allophane originating from the volcanic areas of the Andes Mountains is mixed with limestone or metamorphic rock sediments that contain diffe-rent minerals. Thus the effect of allophane on nutrient availability in these soils is less important than in other geographic locations. The mineralogy of the clay fraction of the Atlantic coast of Costa Rica is variable. However, there are some areas of localized importance where allophane predominates in the clay fraction. This significantly affects the phosphorus concentration in the soil solution, and thus also affects its availability

Table 3. Relevant relationships between the solid and liquid phase in the banana Zones in Latin America.Geographic zone Element or compound Key environmental Known effect on the in the soil solution factors or components root system of the of the soil solid phase banana plant

Alluvial soils of the Restricted Relatively high calcium Nonenorthwest of concentration of P, carbonate contentsHonduras Mg and micronutrients

Atlantic coast of High concentrations of Temporary soil saturation, Poor root developmentGuatemala, Costa soluble Fe and Mn, permanent watertables becoming short and thinRica and Urabá, exchangeable and strongly acid pHColombia aluminum >30%

Pacific coast of Low concentration of P Allophane in the clay NoneGuatemala and fraction of the soilNicaragua

Santa Marta, High soluble salt Very limited rainfall, Weak and injured rootsColombia contents elevation above sea level, poor internal soil drainage, high watertable

Lake of Maracaibo, High concentration of Minerals rich in sodium NoneVenezuela exchangeable sodium in the coarse soil fraction, in soil solution elevation over sea level and climate

Machala, Ecuador Very low concentration Degraded micas and illite of K in the clay fraction


for plant uptake. This condition is more common in the foothills of the central mountain range of the country.

The relatively high concentrations of iron and aluminum oxides and hydroxides in the soils formed in highly weathered sediments also restrict the phosphorus concentra-tions in the soil solution. This condition is common in the soils of the gently rolling hills of Pocosí and La Sierpe in Limon Province of Costa Rica. The cation exchange capacity/clay ratio of subsoil horizons suggests significant amounts of halloysite and possibly kaolinite; thus, given the rainfall regime of the area, nutrient leaching can be significant.

Aluminum, iron and manganese Bananas can be grown in soils with a wide pH range, 5.5 to 7.7 (Stover and Simmonds 1987). Actually, this range is wider - reaching values slightly higher than 8 and lower than 4.7. This also implies a wide range of aluminum and manganese concentrations in the soil solution. The symptoms of iron and manganese toxicities in bananas have been described by Lopez et al. (2001), but they were not associated with a significant decrease in yield or plant development. Similar symptoms have been reported by Norman et al. (1995). Production experiences indicate that in soils with a pH range from strongly to extremely acid, exchangeable aluminum concentrations are greater than 30%, and together with high soil manganese concentrations, yields are reduced, usually to less than 2,000 boxes/ha/year. It has not been clearly established whether this is due to high soil Al, Mn and Fe concentrations or to low Ca, P, K, Mg, Zn and B concentrations, that are usually very low in soils with these characteristics. Short, thin, branched roots and shallow corms are found in soils with different physical characteristics but similar Al, Fe and Mn contents. Microscopic tissue studies related to this condition have not been reported. The response of bananas to liming has been inconsistent, probably due to inadequate rates (soil-incorporated or applied to the soil surface), soil water content and mesh size of the liming materials used. High lime and dolomite applications can alter the Ca/Mg, Mg/K ratios and decrease the availability of phosphorus and micronutrients. It has been suggested that bananas have a certain degree of tolerance to aluminum toxi-city (Lahav and Turner 1992). Production experiences also suggest differences in tole-rance among the commercial banana varieties currently used. Often, banana producers point out that the ‘Valery’ variety shows more resistance than ‘Williams’ and in turn this variety more than ‘Grande naine’ under the soil chemical conditions previously indicated. It is clear that this subject demands more research taking into consideration possible differences among the various banana varieties.

The organic fraction: humic acids Climatic conditions and the different plant species that occupied a given area before cultivation of banana or plantain determine the kinds of organic acids found in the soil. Thus, a wide variety of organic acids can be expected in soils devoted to banana production. The physical and chemical characteristics of humic substances are quite different (Varani and Pinton 2001). The effect of humic substances on the interactions discussed in this work can range from weak to strong depending upon their concentra-tion in the soil system, mainly concerning the exchange reactions between the solid and liquid soil phases. Production experiences indicate that plant responses are different in


soils that have similar morphological, physical and chemical characteristics, including organic matter contents, but that had different plant species before banana cultivation. This explains differences in response to the application of organic matter, humic subs-tances, and crop residues after long fallow periods with different plant species. Organic matter composition also determines the population and kind of soil microorganisms in the rhizhosphere. This issue has been little studied in banana cultivation; however, it is of considerable importance and deserves much more investigation in the future. The chelating role of humic substances has also been widely recognized to offer significant opportunities for future research.

The liquid phaseThe equilibrium between phasesNutrient uptake by banana plants, as any other higher plant, depends on the concentra-tion and activity of the ions in the soil solution. Thus, the equilibrium of the nutrients in the soil solution at the time of plant uptake determines the rate of their uptake (Barber 1995). This situation becomes very complex under field conditions since the soil is an open system subjected to changing imports and exports of matter and energy in the root zone. This prevents the establishment of a long-term equilibrium between the solid and the liquid phase; and, therefore alters the supply of nutrients to the rhizosphere of the banana plant.

The practical applications of this concept can be significant for banana production, con-sidering the fact that the concentration of soluble salts in the soil solution is constantly altered by the addition of fertilizers like KCL, nitrates and sulfates. This effect is not the same throughout the year due to soil water conditions in the various banana zones of Latin America. It is obvious that salt concentration in the soil, due to the addition of fer-tilizers, is less important in rainy areas with uniform rain distribution than in dry zones. This implies that a key aspect of banana plant nutrition in rainy areas is to maintain an adequate nutrient concentration in the soil solution. This highlights the importance of frequent fertilizer applications in banana zones of high and uniform rainfall throughout the year such as the Atlantic coast of Costa Rica, Panama, and Urabá in Colombia.

The climatic factorThe effect of soluble salts generated by fertilizer applications is stronger in areas with well-defined dry and rainy seasons even though the areas may be irrigated. Irrigation systems are usually designed to prevent significant leaching. In these areas, the effect is relieved during the rainy season such as in the North coast of Honduras and the Pacific coast of Nicaragua due to natural leaching of the soil profile. However, in areas where evapotranspiration exceeds rainfall for long periods, the soluble salt concentra-tion generated by fertilizers or weathered salt containing rocks can affect banana yields significantly over time.

Cations in the soil solutionAs previously pointed out, potassium and to a lesser extent magnesium concentrations in the soil system depend to a large extent on the kinds and content of the clay minerals in the soil. The equilibrium reactions are altered by leaching, fertilizer applications, and the kinds and content of humic substances. On the other hand, the sodium ion con-


centration in the soil solution depends on the soil content of albite (sodium feldspars), the cation exchange capacity, rainfall and internal soil drainage. The effect of high sodium concentrations in the soil solution is better understood when it is expressed as the Sodium Absorption Ratio (SAR) and the Exchangeable Sodium Percentage (ESP). In banana production, values greater than 5 and 7%, respectively, limit yields and retard vegetative growth. Under field conditions, banana roots are short and thin and have black or dark brown epidermal lesions when soil sodium is at the concentrations indicated above.

Necrosis along the margin of the leaves and slow growth are better field diagnostic criteria than the symptoms shown by roots. Field observations indicate that the mecha-nisms by which sodium affects banana plants are similar to those found in other sus-ceptible plant species as described by Maas and Nieman (1978) and Hale and Orcutt (1987). Typical cases of this situation occur in the banana zones of Santa Marta in Colombia, South of the Maracaibo Lake in Venezuela, and the Dominican Republic. High soil sodium concentration occurs in low-lying areas influenced by high tides such as on the Pacific Coasts of Mexico, Guatemala, and Machala (Ecuador), where the watertables are high and leaching rates are low. Given how difficult it is to drain the soil effectively and wash sodium out of the root zone in these areas, compounds that may increase the cation exchange capacity of the soil can lower the sodium ion activity in the soil solution substantially. Among these substances are humic acids, highly decom-posed compost and synthetic polymers with high negative charge.

Soluble saltsSoils devoted to banana production with relatively high soluble salt concentrations with or without high sodium contents, occur in areas where evapotranspiration greatly exceeds precipitation most of the year (Santa Marta in Colombia, northern Dominican Republic and the southern Maracaibo Lake area in Venezuela), and soil leaching is limited or where elevation is not high enough above sea level to allow effective soil drainage (several areas on the Pacific coasts of Guatemala, El Salvador and Ecuador). Electrical conductivities above 5 dS/m affect banana production; however, the kind of soluble salt is more important to banana plant yields than the concentration itself. Sodium chloride strongly affects crop production due to the toxic action of both ions. Sodium is especially damaging compared with calcium sulfate, calcium carbonate or magnesium sulfate. The effect caused by calcium carbonate in banana production is more indirect; high soil solution pH strongly affects the availability of phosphorus and most micronutrients and alters the balance among calcium, magnesium and potassium. On the other hand, very high magnesium and sulfate concentrations could potentially cause severe damage to the crop as has been found in other higher plants (Fitter and Hay 1987). However, this condition has not been reported in bananas specifically.

As previously indicated, temporary high soluble salt contents in the rhizosphere can occur as a result of using large amounts of high salinity index fertilizers. Symptoms observed in the field indicate that the effects of soluble salts in banana plants, and the root system in particular, are generally the same as those described by Marschner (1995) for susceptible higher plants. In practice, the improvement of the rhizosphere affected by high soluble salt concentrations depends on the availability of enough good


quality water to wash the soluble salts out of the soil root zone and of a drainage system that will allow leaching to take place.

Soil solution reaction The bioavailability of most nutrients is strongly influenced by the pH of the soil solu-tion and the changes that it undergoes with the application of fertilizers and other com-pounds. The solubility of the various compounds used greatly affects the magnitude of changes in the soil solution pH. Fertilizers that are applied through the irrigation system are more advantageous in lowering the soluble salt concentration than hand-applied fertilizers since the rate of fertilizers applied by hand are usually higher due to the lower application frequency.

Without doubt, the application of fertilizers via the irrigation system improves plant nutrient uptake. It is almost impossible to attain the relative stability of the soil solution when the cation exchange capacity of the soil is strongly pH dependent as is typical of the Andosols of the Pacific coasts of Central America, Martinique and Guadeloupe. The effect of fertilizer additions in these soils make matters more complex as the soil solu-tion pH changes frequently causing the Cation Exchange Capacity (CEC) of the soil to change frequently as well. The almost constant changes in the CEC make the behavior of Andosols less predictable than the non-volcanic soils of other banana producing zones in Latin America. This perhaps partially explains the inconsistent responses to the application of certain nutrients in these soils.

Redox reactions Different states of redox potential occur with the variation of oxygen concentrations in the soil system, as a consequence of fluctuations in the watertable or the occurrence of perched watertables that are very common throughout the banana zones of Latin America. Also, long periods of rainfall associated with cold fronts on the Atlantic Coasts of Guatemala, Honduras and Costa Rica commonly result in saturation of the soil profile and changes in its redox-potential. Changes in the solubility of iron and manganese and their affects on banana production rank among the most important con-sequences of this phenomenon. To what extent this negatively affects the banana root system has not been clearly established, since the consequences of soil saturation are multiple and always negative for the crop. The toxic effects of iron and manganese have not been clearly established since the active uptake of nutrients stops after oxygen has been depleted from the rhizosphere. It is of great commercial importance to establish the differences among cultivars in tolerance to high concentrations of iron and manga-nese. This information may play an important role in planning planting strategies on a regional level.

It is also important to recognize that the temporary saturation of the entire soil profile, due to long and intense periods of rainfall, sometimes negatively affects the banana root system more than permanent and static watertables. Banana roots grow during the dry season in soil horizons that will be saturated in the rainy season when many roots die.

The rhizosphereThis is the zone of interaction in the soil-plant relationship; thus it is the most complex component of the system (Pinton et al. 2001). In particular, this component has been


little studied in bananas. Judging by the level of understanding reached in other species, it can be inferred that it is also vital to banana production that progress be made in innovative technologies to improve the rhizosphere. The banana industry has recently recognized the need to improve the conditions of the rhizosphere. Its physical, chemical and biological components have been improved in the last decade. This has involved a wide spectrum of technologies among which humic substances, the application of fer-tilizers through the irrigation system, the use of hormones, soil physical conditioning, mycorrhizae inoculation, and lately precision agriculture are the most relevant. All these technologies are aimed at improving the conditions of the rhizosphere.

Root system performance has been substantially improved by mycorrhizae inoculations under physically, chemically and biochemically adverse soil conditions. Inoculation with mycorrhizae has resulted in greater uptake of P, K, S, and Cu in the ‘Galil 7’ variety (Solis 2003), and Cu, Mn, and Zn in cv. ‘Williams’ (Medina 2003) under green-house conditions. Faster growth of inoculated plantain plants has been reported under greenhouse conditions, 18 days faster than for the control plants. In the same experi-ment, inoculation of banana roots with mycorrhizae, Mycoral ®, resulted in greater production of dry matter, but not in a faster growth rate. The results of the tests showed a 46% increase in dry weight and a 56% increase in root volume in cv. ‘False horn’ plantain and a 43% increase in dry weight in cv. ‘Williams’ banana. In all cases, there was an increase in the uptake of phosphorus, sulfur, potassium, zinc, copper and iron (Zamorano Biotechnology Unit 2003).

These findings suggest that future advances in similar technologies will allow for a wider range of soils to be planted with bananas at lower production costs under limited soil conditions.

Conclusions and recommendationsThe relationship between soil chemical properties and the banana root system needs more basic and applied research to design more efficient programs to improve rhizos-phere conditions that will result in higher yields and a sustainable soil management system. Basic research should include microscopic tissue analyses, physiology of the banana plant root system as it is related to potentially toxic ions in various soil systems and climates, the effect of humic substances and mycorrhizae in improving rhizosphere performance. On the other hand, applied research should include the effect of chelating agents on nutrient availability for plant uptake, the possible effect of humic substan-ces in reducing the activity of toxic ions in the soil solution, kinds and rates of liming materials, and the relationship between physical, chemical and biological conditions and their effect on crop yield.

ReferencesBarber S.A. 1995. Soil Nutrient Bioavailability: a mechanistic approach. 2nd ed. John Wiley and Sons Inc., New York.

Pp. 41-42.

Fitter A.H. & R.K.M Hay. 1987. Environmental Physiology of Plants. 2nd ed. Academic Press, London. Pp. 227-228.

Hale M.G. & D.M. Orcutt. 1987. The Physiology of Plants under Stress. John Wiley and Sons Inc. New York. Pp. 21-22.

Lahav E. & D.W. Turner. 1992. Fertilización del banano. 2nd ed. Bulletin No. 7. Instituto Internacional de la Potasa. Basilia, Suiza. 33pp.


Lopez A., A. Vargas, J. Espinoza & R. Vargas. 2001. Síntomas de deficiencias nutricionales y otros desordenes fisioló-gicos en el cultivo del banano (Musa AAA), Guía de campo. Potash and Phosphate Institute, Quito, Ecuador.

Maas E.V. & R.H. Nieman. 1978. Physiology of plant tolerance to salinity. Pp. 283-284 in Crop Tolerance to Suboptimal Land Conditions. (G.A. Jung, ed.). American Society of Agronomy: Special Publication No. 32. Madison WI.

Marschner H. 1995. Mineral Nutrition of Higher Plants. Academic Press. London. 662pp.

Medina R.A. 2003. Evaluación de la micorriza vesículo-arbuscular, Mycoral ®, en cormos de banano y plátano en vivero. Undergraduate Thesis. Panamerican School of Agriculture. El Zamorano, Honduras.

Norman M.J.T., C.J. Pearson and P.G.E. Searle. 1995. The Ecology of Tropical Food Crops. Cambridge University Press, Cambridge. 330pp.

Pinton R., Z. Varanini & P. Nannipieri. 2001. The Rhizosphere as a site of biological interactions among soil compo-nents, plants and microorganisms. Pp. 2-3 in The Rhizosphere: biochemistry and organic substances at the soil-plant interface. (R. Pinton, Z. Varanini,p & P. Nannipieri, eds). Marcel and Dekker, Inc. New York.

Solis J.C. 2003. Efecto de la micorriza vesículo-arbuscular, Mycoral ® en vitroplantas de banano y plátano en vivero. Undergraduate Thesis. Panamerican School of Agriculture, El Zamorano, Honduras.

Stover R.H. & N.W. Simmonds. 1987. Bananas. 2nd ed. Longman Scientific & Technical. 468pp.

Zamorano Biotechnology Unit. 2003. Resultados de las investigaciones sobre uso de biofertilizantes micorrizados en producción Agrícola en Honduras, realizadas en la Escuela Agrícola Panamericana El Zamorano. Special Report. Career of science and agricultural production, Panamerican School of Agriculture. El Zamorano, Honduras.

Varanini Z. & R. Pinton. 2001. Direct versus indirect effects of soil humic substances on plant growth. Pp. 142 in The rhizosphere: biochemistry and organic substances at the soil-plant interface. (R. Pinton, Z. Varanini & P. Nannipieri, eds). Marcel Dekker. Inc. New York.

The problem of banana root deterioration and its impact on production142 Banana soil acidification and its relationship with increased aluminium concentrations C.A. Gauggel et al. 143E. Serrano

Banana soil acidification in the Caribbean coast of Costa Rica and its relationship with increased aluminium concentrationsEdgardo Serrano1

AbstractThe relationship among soil acidity, pH and aluminium was studied in plantations of different ages located in the areas east and west of the Reventazón River, Costa Rica. Mini-trenches were dug to study, both physically and chemically, the three horizons in the upper 60 cm of the soil profile in the area under the influence of fertilizers. The surface horizon (Ap) ranged from 8 to 12 cm deep, had accumulated humic organic matter and its chemical and physical properties were more likely to be affected by the applications of agrochemicals. The next horizon (Bw1) ranged from 20 to 30 cm deep, and was characterized by a change in colour, structure and/or texture. The third soil horizon (Bw2) from 28 to 32 cm deep, also differed from the first horizon in colour, texture and/or structure. In the Western zone (n=75), an increasing linear effect was found between crop age, soil acidity and aluminium concentration in the Ap, Bw1 and Bw2 horizons. The soil pH decreased linearly with age of plantation in each of the three horizons. The same effect was observed in the Eastern zone (n=165) but only in the Ap and Bw1 horizons. By sampling the same production units for three crop cycles, a decrease in pH from 6.5 to 5.1 and an increase in acidity from 0.07 to 1.16 cmol(+)/L was observed in the andosols west of the Reventazón river. This acidification was induced by the acid reaction of the fertilizers used. In most of the world’s tropical areas where soils are acid, aluminium toxicity limits crop production.

Resumen - Acidificación de los suelos bananeros en la región Caribe de Costa Rica y su relación con el incremento en las concentraciones de aluminioLa relación entre la acidez, el pH y el Aluminio se estudió en fincas de diferente edad, ubicadas en la zona Este y Oeste del río Reventazón. Se utilizaron mini calicatas para describir química y físicamente los tres horizontes presentes en los primeros 60 centímetros del perfil del suelo, en el área de influencia de los fertilizantes. El horizonte superficial (Ap) osciló entre 8 y 12 cm, acumuló materia orgánica humificada de la cual derivó su color oscuro y sus propiedades físico-químicas fueron alteradas por la adición de agroquímicos. El siguiente horizonte (Bw1) mostró una amplitud de 20 a 30 cm y se caracterizó por un cambio de color, estructura y/o textura. El tercer horizonte (Bw2) osciló entre 28 a 32 cm y se separó del anterior por un cambio nuevamente en el color, estructura y/o textura. En la zona Oeste (n =75) se encontró un efecto lineal creciente entre la edad del cultivo y la acidez con la concentración de aluminio en los horizontes Ap, Bw1 y Bw2. El pH del suelo decreció linealmente con el tiempo en cada uno de los 3 horizontes. El mismo efecto se observó en la zona Este (n = 165) pero solo en los horizontes Ap y Bw1. Muestreando las mismas unidades de producción durante tres ciclos de cultivo, en andisoles de la zona oeste del río Reventazón, se encontró que el pH en la banda de fertilización disminuyó de 6,48 a 5,07 y la acidez aumentó de 0,07 a 1,16 Cmol(+)/L. Esta acidificación fue inducida por el efecto de los fertilizantes de reacción ácida que se usan en la actividad. En la mayor parte de las áreas tropicales del mundo donde los suelos son ácidos, la toxicidad por aluminio limita la productividad del cultivo.

IntroductionThe banana root system is affected by abiotic and biotic factors. Among the biotic factors are plant parasite nematodes, opportunistic fungi that cause root rot, micro-

1 Dirección de Investigaciones, CORBANA. Apdo. 390-7210 Guápiles, Costa Rica. e-mail: [email protected]


organisms associated with the accelerated biodegradation of nematicides, and foliage pests (insects and fungi) that damage leaves at flowering and harvest, affecting the root system balance. Within the abiotic factors there are water excess and deficits, extreme temperatures (low and high), low incidence of photosynthetically active radiation, soil compression, poor drainage (internal and superficial), and soil acidification which results in toxic concentrations of soluble Al+3 (Avilán et al. 1982, Sancho 1993, Serrano and Marín 1998, López and Espinosa 2000, Serrano 2002, Moens et al. 2003).

Al+3 is not considered a nutrient for banana since it is not essential for the plant to complete its life cycle (Lahav and Turner 1992). The problem lies in the possible toxic effects of soluble Al-3 created in the soil when increased amounts of ammonium based fertilizers decrease soil pH (López and Espinosa 2000, Serrano 2002). In studies with acid nutrient solutions (78 µM Al), Rufyikiri et al. (2000, 2001) observed a decrease in total plant dry weight, in the number and diameter of roots, in the Ca and Mg con-tents, and in the total water and nutrient absorption of ‘Grand naine’ plants. Rodríguez et al. (1985) classified the ‘Grande naine’ banana cultivar and ‘Maricongo’ plantains as tolerant to a high Al+3 saturation. Nevertheless, these results are difficult to transfer directly to the banana soils of the Caribbean coast of Costa Rica. Rodriguez’s results were obtained in ultisols and oxisols; orders of acid soils that practically do not exist in the alluvial zones of Costa Rica that are used to cultivate bananas. The majority of soils in Costa Rica’s Caribbean Zone are andosols and inceptisols, that are not naturally acid (López 1963, Jiménez 1972, López and Solís 1991, Nieuwenhuyse et al. 1993, Stoorvogel and Eppink 1995, Mata 2000).

This paper examines acidity induced in the fertilization band, pH and Al+3 in the culti-vated banana soils in Costa Rica’s Caribbean coast.

Main sources of soil acidity There are various natural processes and crop management procedures (Thomas and Hargrove 1984, Lora 1994, Espinosa and Molina 1999, López and Espinosa 2000, Matsumoto, 2002) that increase soil acidity and decrease soil pH.

Among the most important are:

a) Nutrient removal. This occurs when the crop roots absorb cations (K+, Ca+2 and Mg+2) and release H+, to maintain ionic balance.

b) Nitrogen based fertilizers. When nitrogen fertilizers that either contain or form ammonium (NH4

+) are used, the ammonium changes to nitrate by means of nitrification and produces an excess of H+ that acidifies the soil (Equations 1 and 2).

2NH4+ + 3O2 2NO2- + 4H+ + H2O (1)

Nitrosomes

2NH4+ + O2 2NO3- (2)

Nitrobacter

The soils where bananas are grown are intensively fertilized with N. Between 375 and 450 kg of N/ha/yr are applied to satisfy the crop need, usually using ammonium based


fertilizers that, to a greater or lesser degree, acidify the soil. This residual or induced acidity is measured through the physiological acidity index, which indicates the amount of CaCO3 that could neutralize the acidity caused by 100 kg of fertilizer. The higher the net value of this index, the greater the residual effect that the fertilizer produces (Table 1). To supply 387 kg of N/ha/yr on the farms east of the Reventazón River, 1402 kg of raw material is used, of which 81% are nitrogen sources that either have or produce ammonium (Table 1).

Table 1. Sources of nitrogen fertilizers used in banana plantations of the zone east of the Reventazón river. Nitrogen fertilizers N Fertilizer Physiological kg ha-1 yr-1 kg ha-1 yr-1 acidity index*

Ammonium sulfate (NH4)2SO4 72 342 -112

Ammonium nitrate (NH4)2NO3 130 388 -63

Urea (NH4)2CO 92 200 -84

Diammonium phosphate [(NH4)2HPO4] 37 203 -74

Calcareous ammonium nitrate 57 269 0

Total 388 1 402 *Physiological acidity index = (-) kg CaCO3/100 kg of fertilizer.

c) Organic matter mineralization. This is a natural process that produces NH4+ as a

final product of organic matter decomposition. This NH4+ source also acidifies the soil,

although more slowly than the addition of ammonium based fertilizers.

d) Soluble aluminium hydrolysis. Aluminium is the third most abundant metal in the earth’s crust, and it exists in the soil in an insoluble form as aluminosilicates and oxides. When fertilizers that are applied to the soil, lower the pH to less than 5.3, a soluble form of Al exists as octahydro-hexahydrated [Al(H2O)6

+3], frequently abbreviated to Al+3. Hydrolysis occurs when the charge/size ratio of the cation is sufficiently large to break the H-O links. The result is hydrate ionization which produces hydrogen ions. Reactions (3), (4) and (5) were abbreviated and are as follows:

Al+3 + H2O Al(OH)2+ + H+ (3)

Al(OH)2+ + H2 Al(OH)2+ + H+ (4)

Al(Oh)+ + H2O Al(OH)30 + H+ (5)

Each reaction releases H+ and contributes to soil acidification. This acidity increase promotes the presence of Al+3 which is ready to react again. According to the charge/size ratio of the different soil cations, the affinity for hydrolysis that causes acidification is as follows, in decreasing order: Al+3(44)>Fe+3(36)>Mg+2(14)>Cu+2(14)>Zn+2(14)>Mn+2(13)>Ca+2(10)>Na+(2).


Sampling methodologyA total of 14 farms located in the alluvial plain of the Caribbean coast of Costa Rica were sampled: 5 farms of different ages west of the Reventazón River, distributed among Sarapiquí, Guácimo and Pocosí counties; and 9 farms east of the Reventazón River in Siquirres, Matina, Limón and Talamanca counties. Seventy five mini-trenches were dug in the zone west of the river on farms that had been planted for 1, 10, 12 or 30 years, while 165 mini-trenches were dug on the eastern zone on farms that had been planted for 1, 3, 4, 7, 10, 11, 17 or 30 years.

The sampling was done following the methodology described by Jaramillo and Vasquez (1990), which consists in excavating a 60 cm wide x 60 cm long x 60 cm deep ditch (mini-trench) in the fertilization band of recently flowered banana plants. In each mini-trench at least three soil horizons (Ap, Bw1 and Bw2 ) were studied. The surface horizon (Ap) ranged from 8 to 12 cm deep, had accumulated humic organic matter, from which its dark colour was derived, and its chemical and physical properties were more likely to be affected by the applications of agrochemicals. The next horizon (Bw1) ranged from 20 to 30 cm deep, and was characterized by a change in colour, structure and/or texture. The third soil horizon (Bw2), from 28 to 32 cm deep, also differed from the first horizon in colour, texture and/or structure (Figure 1). To measure soil pH, exchangeable acidity and Al+3, samples were taken from each horizon. The pH was determined in water using the McLean (1982) procedure with a 1:2.5 soil:water ratio. The readings were taken with an Orion Research 701 A potentiometer. To determine exchangeable acidity, 1N KCl was used as the extracting solution, titrating it with NaOH according to the procedure described by Thomas (1982).

The effect of plantation age on soil acidificationIt was observed on the western farms (n = 75) that as plantation age increased, the acidity and the Al+3 increased linearly in the Ap, Bw1 and Bw2 horizons (Table 2). The pH decreased linearly in all three horizons. Similarly, it was observed that as plantation (n = 165) age increased in the eastern zone, the acidity and the aluminium increased and

Figure 1. Sampling methodology based on mini-trenches.


pH decreased, in a linear manner but only in the Ap and Bw1 horizons (Table 3). The acidity in the deeper horizon (Bw2) of the western soils could be related to a greater leaching of the nitrogenous fertilizers towards the deeper horizons. The western zone has soils that are moderately deep, with medium and moderately fine textures, and a greater infiltration rate and hydraulic conductivity than the eastern zone soils (Serrano and Marín 1998, Alvarado et al. 2001). Additionally, in the period between 1987 and 2002 the annual average precipitation on the western side of the Reventazón river was 4152 mm in comparison with 3547 mm on the eastern side.

Sampling the same Andisol production units west of the Reventazón river during three crop cycles, it was observed that pH in the fertilization band decreased from 6.48 to 5.07 and acidity increased from 0.07 to 1.16 cmol(+)/L, with a consequent increase of Al+3 from 0.02 to 0.99 cmol(+)/L in a two year period. The induced acidification

Table 2. Relationship between plantation age and soil acidity, pH and aluminium in three soil horizons in banana plantations of the zone west of the Reventazón river (n = 75). Variable Horizon Plantation age (years) Linear effect*

1 10 12 30

Ap 5.83 5.00 4.78 4.26 dec.**

pH Bw1 6.29 5.31 5.33 4.76 dec.**

Bw2 6.21 5.81 5.54 5.21 dec.**

Exchangeable Ap 0.70 1.76 1.64 4.35 inc.**

acidity Bw1 0.19 1.45 0.24 4.39 inc.**

cmol(+)/L Bw2 0.12 0.58 0.14 4.72 inc.**

Ap 0.64 1.01 0.89 2.72 inc.**

Al Bw1 0.17 0.81 0.04 2.49 inc.**

cmol(+)/L Bw2 0.11 0.28 0.02 2.76 inc.*** The linear effect is the significance of the linear regression between the variable and plantation age. ** P<0.01. dec = negative slope, inc = positive slope.

Table 3. Relationship between plantation age and pH, soil acidity, and aluminium in three soil horizons of banana plantations located in the zone east of the Reventazón river (n = 165). Variable Horizon Plantation age (years) Linear

1 3 4 7 10 11 17 30 effect*

Ap 5.50 5.25 5.08 4.26 4.95 4.45 4.38 4.41 dec.**

pH Bw1 6.41 5.94 6.01 5.63 5.83 5.50 5.27 4.96 dec.**

Bw2 6.63 6.35 6.14 5.84 6.15 5.96 5.96 5.25 N.S.

Exchangeable Ap 1.57 3.55 0.79 2.41 3.73 4.70 2.21 3.96 inc.**

acidity Bw1 0.34 0.26 0.17 0.25 0.63 1.18 1.03 2.11 inc.**

cmol(+)/L Bw2 0.34 0.13 0.09 0.16 0.48 0.95 0.20 1.08 N.S.

Ap 0.85 1.65 0.40 1.12 2.05 2.27 0.55 4.69 inc.**

Al Bw1 0.18 0.09 0.05 0.07 0.23 0.55 0.75 2.45 inc.**

cmol(+)/L Bw2 0.09 0.00 0.00 0.00 0.30 0.62 0.14 0.33 N.S.* The linear effect is the significance of the linear regression between the variable and plantation age. ** P<0.01, N.S.= not significant, dec = negative slope, inc = positive slope.


is cumulative in the fertilizer application band, which is consistent with the increase in plantation age. When pH decreases to 5.3, the Al+3 becomes soluble and as a direct consequence of this there are fewer roots and less water and nutrient absorption by the banana crop (Lora 1994, Rufyikiri et al. 2000, 2001).

The soils and drainage section of CORBANA in Costa Rica, recommends some crop management practices to diminish the impact of induced acidification, such as:

• Application of calcium amendments as calcite (calcium carbonate) and dolomite (calcium carbonate and magnesium) to precipitate Al+3 and to contribute Ca and Mg.

• Application of decomposed or semi-decomposed organic matter to adsorb Al+3.

• Application of fertilizers in lesser quantities but more often (increasing from 17 to 26 cycles of fertilizer application/yr).

• Application of fertilizer in a 90 cm radius around the follower.

ReferencesAlvarado A., F. Bertch, E. Bornemisza, G. Cabalceta, W. Forsythe, C. Henríquez, R. Mata, E. Molina & R. Salas. 2001.

Suelos derivados de cenizas volcánicas (Andisoles) de Costa Rica. Asociación Costarricense de la Ciencia del Suelo, San José, Costa Rica. 112pp.

Avilán L., L. Meneses & S. Rómulo. 1982. Distribución radical del banano bajo diferentes sistemas de manejo de suelos. Fruits 37:103-110.

Espinosa J. & E. Molina. 1999. Acidez y encalado de los suelos. Instituto de la Potasa y el Fósforo, Quito, Ecuador. 42pp.

INPOFOS. 1997. Manual Internacional de Fertilidad de Suelos. Potash & Phosphate Institute, GA, USA. 72pp.

Jaramillo R. & A. Vásquez. 1990. Manual de procedimientos para la presentación y realización de estudios detallados de suelos y clasificación de tierras para el cultivo de banano. San José, Costa Rica, Corporación Bananera Nacional. 29pp.

Jiménez T. 1972. Génesis, clasificación y capacidad de uso de algunos suelos de la Región Atlántica de Costa Rica. Tesis Ing. Agr. San José, Costa Rica, Facultad de Agronomía, Universidad de Costa Rica. 180pp.

Lahav E. & D.W. Turner. 1992. Fertilización del banano para rendimientos altos. Segunda edición. Boletín No. 7. Instituto de la Potasa y el Fósforo, Quito, Ecuador. 71pp.

López A. & P. Solís. 1991. Contenidos e interacciones de los nutrimentos en tres zonas bananeras de Costa Rica. Corbana 15:25-32.

López A. & J. Espinosa. 2000. Manual on the nutrition and fertilization of banana. National Banana Corporation, Limón, Costa Rica and Potash & Phosphate Institute Office for Latin America, Quito, Ecuador. 57pp.

Lora R. 1994. Factores que afectan la disponibilidad de nutrimentos para las plantas. Pp. 28-55 in Fertilidad de Suelos: Diagnóstico y Control. (Silva F., ed.). Sociedad Colombiana de la Ciencia del Suelo. Bogotá, Colombia.

López C. 1963. Identificación y clasificación de los minerales de arcilla presentes en nueve suelos de Costa Rica. Tesis Ing. Agr. Facultad de Agronomía, Universidad de Costa Rica, San José, Costa Rica. 57pp.

Mata R. 2000. Caracterización de suelos: Proyecto Hidroeléctrico Jiménez. Tecnoambiente Centroamericano. San José, Costa Rica. 92pp.

Matsumoto H. 2002. Plant roots under aluminium stress. Toxicity and tolerance. Pp. 821-838 in Plants Roots: The Hidden Half. 3rd ed. (Y. Weisel, A. Eshel & U. Kafkafi, eds). Dekker, New York, USA.

McLean E.O. 1982. Soil pH and lime requirement. Pp. 199-224 in Methods of Soil Analysis. Chemical and Microbiological Properties. (A.L. Page R.H. Miller & R. Keeney, eds). ASA. Agronomy Series No. 9, Madison, Wisconsin, USA.

Moens T., M. Araya, R. Swennen & D. De Waele. 2003. Biodegradación acelerada de nematicidas después de apli-caciones repetidas en una plantación comercial de banano. Pp. 35-36 in Manejo convencional y alternativo de la Sigatoka negra, nematodos y otras plagas asociadas al cultivo de Musáceas (G. Rivas, ed.). INIBAP, MUSALAC and FUNDAGRO, Guayaquil, Ecuador.

The problem of banana root deterioration and its impact on production148 Banana soil acidification and its relationship with increased aluminium concentrations C.A. Gauggel et al. 149T. Pattison et al.

Nieuwenhuyse A., A. Jongmans & N. Van Breemen. 1993. Andisol formation in Holocene beach ridche plain under the humid climate of the Atlantic coast of Costa Rica. Geoderma 57:423-442.

Rodríguez J., E. Rivera & F. Abruna. 1985. Crop response to soil acidity factors in ultisols and oxisols in Puerto Rico. 14. Plantains and bananas. Journal of Agriculture of the University of Puerto Rico 69:377-382.

Rufyikiri G., D. Nootens, J.E. Dufey & B. Delvaux. 2000. Effect of aluminium on bananas (Musa spp.) cultivated in acid solutions. 1. Plant growth and chemical composition. Fruits 55:367-378.

Rufyikiri G., D. Nootens, J.E. Dufey & B. Delvaux. 2001. Effect of aluminium on bananas (Musa spp.) cultivated in acid solutions. 2. Water and nutrient uptake. Fruits 56:5-15.

Sancho H. 1993. Distribución radicular del clon Valery (Musa AAA Cavendish Gigante) en tres diferentes condiciones de drenaje. Pp. 32-34 in Informe Anual Departamento de Investigaciones y Diversificación Agrícola. CORBANA S.A. San José, Costa Rica.


Serrano E. 2002. Diagnóstico físico-químico del suelo y su relación con el crecimiento del cultivo del banano en fincas con diferentes condiciones edafoclimáticas de la zona Caribe de Costa Rica. Pp. 113-117 in Informe anual 2002, Dirección de Investigaciones (Sandoval, J. ed.). CORBANA, San José, Costa Rica.

Stoorvogel J. & G.P. Eppink. 1995. Atlas de la Zona Atlántica Norte de Costa Rica. Programa Zona Atlántica (CATIE/UAV/MAG), Guápiles, Costa Rica.

Thomas G.W. & W.L. Hargrove. 1984. The chemistry of soil acidity. Pp. 3-56 in Soil Acidity and Liming. (F. Adams, ed.). ASA-CSSA-SSSA. Wisconsin, USA.

Thomas G.W. 1982. Exchangeable cations. Pp. 159-165 in Methods of Soil Analysis. Chemical and Microbiological Properties. 2nd ed. (A.L. Page, R.H. Miller & R. Keeney, eds). ASA. Agronomy Series No. 9, Madison, Wisconsin USA.

The problem of banana root deterioration and its impact on production148 Banana soil acidification and its relationship with increased aluminium concentrations C.A. Gauggel et al. 149T. Pattison et al.

Banana root and soil health project – AustraliaTony Pattison1,5, Linda Smith2, Philip Moody3, John Armour4, Kim Badcock1, Jenny Cobon2, Velupillai Rasiah4, Stewart Lindsay1 and Lisa Gulino2

AbstractThe banana plant forms an adventitious root system that is dependent on soil physical, chemical and biological properties to function efficiently. A pot experiment demonstrated that increasing soil compaction was able to significantly reduce the weight of banana roots and shoots. However, in the presence of Radopholus similis the effects of soil compaction were obscured, due to the significant reduction in root weight caused by the nematode.

The use of a basic set of soil quality indicators that can be readily used by farmers, was linked to soil nematode indicators to determine relationships between soil properties. In a survey of banana fields in North Queensland, different diameter root classes were affected differently by changing soil properties. Banana roots greater than 5 mm diameter were positively correlated with aggregate stability and negatively correlated with soil bulk density. Banana roots less than 1 mm were positively correlated with electrical conductivity. Specific interactions between soil properties become apparent as crop production systems become more uniform. This allows farmers to prioritise management options to improve the most deficient soil health indicators.

The addition of organic amendments is one possible method of correcting degrading soils. The use of amendments with high carbon contents, such as grass hay, banana trash and alfalfa hay, were able to significantly suppress R. similis in the roots of banana plants relative to untreated soil. Due to banana production being located near environmentally sensitive areas there is an increasing need to monitor and modify soil management practices. However, this needs to be linked with a framework that allows the integration of all soil components with a system to allow continual improvement in soil management to allow banana production to have minimal impact on the surrounding environment.

Resumen - Proyecto sobre salud del suelo y raíces de banano - AustraliaLa planta de banano forma un sistema radical adventicio que depende de las propiedades físicas, químicas y biológicas del suelo para su buen funcionamiento. Un experimento en macetas demostró que el aumento de la compactación del suelo fue capaz de reducir significativamente el peso de las raíces del banano y de los rebrotes. Sin embargo, en presencia de Radopholus similis los efectos de la compactación fueron dudosos debido a la significante reducción en el peso radicular causada por los nematodos.

El uso de un grupo básico de indicadores de calidad del suelo, que puede ser fácilmente usado por los agricultores, fue correlacionado a indicadores de nematodos en el suelo para determinar las relaciones existentes entre las propiedades del suelo.

En un diagnóstico de suelos bananeros en el norte de Queensland, diferentes clases de diámetro de raíces fueron afectadas diferentemente al cambiar las propiedades del suelo. Raíces de banano con diámetros mayores a 5 mm fueron correlacionadas positivamente con la estabilidad de los agregados del suelo y negativamente correlacionadas con la densidad total. Raíces de banano menores de 1 mm

1 Department of Primary Industries, PO Box 20, South Johnstone, Qld, Australia. 2 Department of Primary Industries, 80 Meiers Rd, Indooroopilly, Qld 4068, Australia. 3 Department of Natural resources and Mines, 80 Meiers Rd, Indooroopilly, Qld 4068, Australia. 4 Department of Natural resources and Mines. PO Box 1054, Mareeba, Qld, Australia. 5 Corresponding author e-mail: [email protected]. Tel. +61 7 4064 1130

The problem of banana root deterioration and its impact on production150 Banana root and soil health project - Australia C.A. Gauggel et al. 151T. Pattison et al.

fueron correlacionadas positivamente con la conductividad eléctrica. Interacciones específicas entre las propiedades del suelo se hicieron más aparentes a medida que los sistemas productivos de los cultivos fueron más uniformes. Esto permite a los agricultores priorizar las opciones de manejo, para mejorar los indicadores de salud del suelo más deficientes. La aplicación de enmiendas orgánicas es un posible método para corregir suelos degradados. El uso de enmiendas con alto contenido de carbono, tales como heno de pasto, residuos de banano y heno de alfalfa, fueron capaces de suprimir significativamente a R. similis en las raíces de las plantas de banano en comparación con suelos no tratados. Debido a que la producción de bananos está localizada cerca de áreas ambientalmente sensibles, existe una necesidad creciente de monitorear y modificar las prácticas de manejo de suelos. Sin embargo, esto necesita estar ligado a un marco de referencia que permita la integración de todos los componentes del suelos en un sistema que permita el continuo mejoramiento del manejo de suelos para asegurar que la producción bananera tenga un impacto mínimo en el ambiente que le rodea.

IntroductionThe banana plant forms an adventitious root system that is wide spreading, unbranched, shallow and gives rise to a dense mat (Blake 1969, Price 1995). The function of banana roots is to provide anchorage, absorption of water and nutrients and synthesis of some plant hormones (Price 1995). The growth of roots is an important factor contributing to banana productivity and is dependent on soil physical, chemical and biological pro-perties (Delvaux 1995). It is suggested that increased size and altered architecture may improve both resource acquisition and anchorage of banana plants, creating a more efficient root system (Price 1995). The condition of the soil has large impact on the development of the root system, with heavy or compacted soils restricting root growth (Araya and Blanco 2001, Price 1995). There are also biological constraints to the deve-lopment and function of the banana root system such as Radopholus similis (Gowen and Queneherve 1990, Gowen 1995, Fogain 2001, Moens et al. 2001).

Delvaux (1995) suggested that soil fertility (health), was a poorly defined concept that not only relied on soil chemical, physical and biological properties, and their interaction with the plant community, but on management practices, farming skills and economics. Doran and Parkin (1996) defined soil health as “the capacity of a soil to function within an ecosystem and land use boundary, to sustain biological productivity, maintain envi-ronmental quality and promote plant and animal health”. Van Bruggen and Semenov (2000) suggested a healthy soil is a stable soil with resilience to stress, high biological diversity and internal cycling of high amounts of nutrients. Knowledge of the function of the soil ecosystem is a basic requirement for soil stewardship (Ferris et al. 2001).

Nematodes are a component of the soil ecosystem that interacts with biotic and abiotic soil factors (Yeates 1979). Because of this interaction, nematodes are excellent bio-indi-cators of soil health, because they form a dominant group of organisms in all soil types, have high abundance, high biodiversity and play an important role in recycling within the soil (Neher 2001, Schloter et al. 2003). Nematodes are heterotrophs, higher in the food chain than micro-organisms and so serve as integrators of soil properties related to their food source, predators and parasites (Ferris et al. 2001, Neher 2001). Nematode diversity tends to be greatest in ecosystems with the least disturbance (Yeates 1999). The disturbance to the soil by environmental or land management practices changes the composition of nematodes (Bongers 1990, Yeates and Bongers 1999, Ferris et al. 2001). There are a number of indices derived from nematode community analysis that


can be used to determine the impact of management changes on the soil ecosystem (Bongers 1990, Yeates and Bongers 1999, Ferris et al. 2001).

However, the use of nematodes as indicators of soil ecosystem health and banana management is not a practical tool for farmers, as it requires specialised knowledge and equipment (Neher 2001). Doran (2002) suggested linking “science to practice” in assessing the sustainability of land management practices, by the use of simple indica-tors of soil quality and health that have meaning for farmers. To embrace changes in environmental management of their land, farmers need to understand why they need to change (Marsh 1998). The best way to achieve this is by the use of participatory research strategies using simple on-farm techniques (Freebairn and King 2003, Lobry de Bruyn and Abbey 2003). A basic set of soil quality indicators was developed by J.W. Doran (USDA-ARS, Lincoln, NE), and developed into an on farm test kit (http://soils.usda.gov/sqi/soil_quality/assessment/kit2.html). The basic set of soil para-meters has been used to measure the effects of changes in soil management on agricul-tural crops (Sarrantonio et al. 1996, Stamatiadis et al. 1999) but not on bananas.

If advances are to be made in improving the efficiency of the banana root system there needs to be a link between indicators and more complex soil processes. Nematodes as integrators of soil properties, linked with chemical, physical and biological indica-tors may increase our understanding of soil management to improve the efficiency of the banana root system. The Banana Root And Soil Health (BRASH) project aims to highlight for farmers limits of the present banana soil management techniques, provide tools and indicators to measure soil health and potential solutions to overcome soil limitations to create a more efficient banana root system.

Materials and methodsImpact of soil compaction A pot experiment was designed to determine the impact of soil compaction on the roots of banana plants (Musa AAA cv. ‘Williams’). The density of soil in 10 L plastic pots was adjusted by adding perlite to field soil to achieve a bulk density of 1.00 and 1.25 g/cm3 or compacted by compressing moist field soil in the pots around a banana plant to achieve a bulk density of 1.50 and 1.75 g/cm3. Half of the plants were inocu-lated with 2390 motile R. similis, 7 days after planting. Each treatment was replicated 8 times in a completely randomized design. The plants were maintained in the glass-house at 20-30oC and were watered by misting twice daily for 15 minutes. The banana plants were fertilized by adding 5 g of Osmocote Plus Mini (16.0:3.5:9.1 + trace elements, Scotts Australia Pty Ltd) at planting.

The banana plants were harvested after 12 weeks. Plant growth parameters, including shoot dry weight, were measured by drying the shoots at 70oC for three days at harvest. A transverse section through the soil next to the plant was made and the number of roots greater than 2 mm in diameter originating from the corm were counted. The roots were then washed free of soil and the fresh root weight determined for different root diameter classes, > 5 mm, 1-5 mm and < 1 mm diameter (Araya and Blanco 2001). Nematodes were extracted from the root system in a misting cabinet for seven days (Hooper 1986) and quantified under a compound microscope.


Banana soil health surveySoil health measurementsA soil health survey method was developed using basic soil quality indicators (Sarrantonio et al. 1996, http://soils.usda.gov/sqi/soil_quality/assessment/kit2.html). On site evaluation of water infiltration was measured with aluminium infiltration rings (150 mm d x 125 mm) inserted 75 mm into the soil. The ring was placed 100 mm from the follower sucker of the banana plant, so that it was in line with the follower sucker and the mother plant. Soil respiration was measured from the head space of the covered infiltration rings for 30 min using CO2 gas sampling tubes (0.01-2.6%, 126SA, Kitigawa, Japan) before determining water infiltration rates. The ring was then inserted level with the soil surface and excavated. All roots within the ring were washed free of soil and banana roots were divided into classes, > 5 mm, 1-5 mm and < 1 mm diameter (Araya and Blanco 2001) and expressed as the weight of root in one litre (dm3) of soil.

A second aluminium ring (75 mm d x 125 mm) was inserted into the soil until level with the soil surface. The ring was excavated from the soil and the soil within the ring was used in the laboratory determination of soil bulk density, gravimetric water content and aggregate stability (Sarrantonio et al. 1996). Additionally, soil pH and electrical conductivity (EC), were determined in a 1:1 soil-water mixture. Soil nitrate-N was determined from the filtered extract of the soil water mixture using test strips (Aquacheck, Hach company, Loveland, USA). A 200 g sub-sample was used for nematode community analyses by placing the sub-sample on a single ply of tissue in water for 1 day (Whitehead and Hemming 1965). The community structure was deter-mined using the method described by Yeates (2001). The nematodes were further sub-divided into functional guilds and nematode indices calculated (Yeates and Bongers 1999, Ferris et al. 2001).

Further observations made in the field included soil temperature (100 mm depth), pre-sence or absence of earthworms, finger number per bunch (Turner et al. 1988), leaf emergence rate and height increase of the follower sucker and the amount of banana trash on the soil surface around the banana plant.

Paired site surveyA survey compared soil and plant measurements (described above) made around bana-na plants with undeveloped plant systems, such as rainforest and pasture. Four sites in north Queensland were compared, located at East Palmerston, Mission Beach, Tully and Kennedy. The sites represented the major banana growing regions and soil types used in banana production in north Queensland. An organic farm was also included in the survey at the Mission Beach site. At each site, five samples were taken from each plant system.

Banana farm surveyTwenty-one banana fields were surveyed using the methods described above, in the major banana production zone of north Queensland. At all sites sampled, banana farms were more than three years old and the cultivars were members of the Cavendish sub-group (Musa AAA). In each farm, five samples were collected.


Banana field surveyAn eight year old banana field (Musa AAA cv. ‘Williams’) near Innisfail, north Queensland was surveyed nine times beginning in October 2002. The sampling occur-red monthly and five banana plants were sampled on each occasion.

Soil amendmentsA pot experiment was established in 10 L plastic containers using three soils from different banana growing areas of north Queensland. The soils were amended with the equivalent of 40 tonnes/ha of grass hay, alfalfa hay, banana trash, mill mud, mill ash (by-products of sugar cane processing), paunch (cattle rumen contents), bio-solid, or municipal waste compost (MW compost). Additional treatments of 120 t/ha of mill ash, 5 t/ha of wolastonite (calcium silicate) and 300 L/ha of molasses were compared with an untreated control. The chemical composition of the amendments was determined (Table 1).

A banana plant (Musa AAA cv. ‘Williams’) was grown in each pot. Half of the pots were inoculated with 860 motile R. similis 3 days after planting the bananas. All treatments were replicated four times in a randomized block design. The plants were maintained in the glasshouse at 20-30oC and were watered by misting twice daily for 15 minutes. Trays were placed at the base of the pots and water collected in the trays returned to the soil surface. The banana plants were fertilized by adding 5 g of Osmocote Plus Mini at planting.

The plants were harvested after 12 to 15 weeks from planting. The shoot dry weight was determined as described previously. The fresh root weight was determined by washing the soil from the roots and allowing the roots to drain on absorbent paper. Nematodes were then extracted from the root system in a misting cabinet for seven days (Hooper 1986) and quantified under a compound microscope.

StatisticsNematode numbers from pot experiments were transformed, ln(x+1), before conduc-ting analysis of variance. Means were separated using the least significant difference (LSD) method and presented as back transformed values. Linear correlations between soil properties were determined for the three surveys. Soil health properties that were significantly correlated (P<0.05) were linked using a concept map. All statistical ana-lyses were performed using Genstat 6 (version 6.1.0.205)

ResultsImpact of soil compactionThe weight of banana roots and shoots declined significantly as the soil bulk density increased (Figure 1). However, the addition of a disease-causing organism such as R. similis obscured the effects of soil compaction causing a significant (P<0.05) reduc-tion in the weight of banana roots (Figure 1). The difference in root weight caused by R. similis was greater at lower than higher bulk densities (Figure 1). There was a signifi-cant decline in shoot weight of bananas with increasing soil bulk density in the absence of R. similis. However, when R. similis was present there was no significant relationship


between shoot weight and soil bulk density (Figure 1). The weight of roots less than 5 mm in diameter were reduced with increasing soil bulk density (Table 2). Similarly, the number of roots greater than 2 mm in diameter originating from the corm was reduced with increasing soil bulk density (Table 2).

Banana soil health

Paired site surveyBanana soils tended to be dominated by plant-parasitic nematodes relative to undeve-loped plant systems (Table 3). This contributed to the lower nematode diversity (H′) in the soil (Figure 2). Aggregate stability tended to be lower on banana producing soils relative to soils in the undeveloped plant systems (Table 3). Aggregate stability was negatively correlated with the amount of nitrate-N in the soil (Figure 2), with banana farms tending to have high soil nitrate-N relative to undeveloped sites, except at the Mission Beach site (Table 3). Electrical conductivity (EC) appeared to be a central indicator of many soil properties (Figure 2). Higher EC was evident on banana farms relative to undeveloped sites, except at Mission Beach (Table 3). Lower EC favoured fungal domination in the soil with nutrients being decomposed by fungal pathways, determined by the channel index (determined by the nematode community analyses as a resource decomposition pathway, Ferris et al. 2001)(Figure 2).

Banana farm surveyThe soil properties on the banana farms exhibited a greater number of significant relationships than comparing different plant systems (Figures 2 and 3). Again, the soil nematode community on banana farms was dominated by plant-parasitic nematodes and low nematode diversity (Table 4). The proportion of plant-parasitic nematodes in the soil was negatively correlated with other nematode feeding types, nematode diver-sity and increased the nematode dominance index (Figure 3).

R2 = 0.98 P<0.05

R2 = 0.88 P<0.05

R2 = 0.96 P<0.05

R2 = 0.01 n.s.

0

20

40

60

80

100

120

1 1.2 1.4 1.6 1.8 2

Soil density (g/cm3)

Root

freshweight(g)

-nematodes+nematodes

A B

0

5

10

15

20

25

1 1.2 1.4 1.6 1.8 2

Soil density (g/cm3)Shoo

tdryweight(g)

+nematodes-nematodes

Figure 1. Effects of increasing soil density on banana root fresh weight (A) and shoot dry weight (B) in the presence and absence of R. similis.


Tabl

e 2.

Ban

ana

root

mea

sure

men

ts i

n so

il at

dif

fere

nt s

oil

bulk

den

siti

es.

Soil

bulk

Fr

esh

root

wei

ght

(g/L

) Co

rm r

oot

num

ber

den

sity

>2

mm

To

tal

root

s >5

mm

1-

5 m

m

<1 m

m (

g/cm

3 )

1.

00

60.0

c

0.3

n.s.

30

.2

b 29

.4

b 5.

1

bc

1.

25

54.9

bc

0.

9 n.

s.

27.8

b

26.2

b

5.8

c

1.

50

42.4

ab

0.

1 n.

s.

19.8

a

22.5

ab

4.

0

b

1.

75

33.9

a

0.3

n.s.

17

.3

a 16

.2

a 2.

1

aM

eans

wit

h th

e sa

me

subs

crip

t w

ithi

n a

colu

mn

are

not

sign

ific

antl

y di

ffer

ent

at t

he 5

% le

vel.

Val

ues

are

the

mea

ns o

f 16

pla

nts

wit

h an

d w

itho

ut R

. si

mil

is.

Tabl

e 1.

Che

mic

al c

ompo

siti

on o

f am

endm

ents

app

lied

to s

oil

befo

re g

row

ing

bana

nas

in p

ots.

Soil

amen

dmen

t N

utri

ent

(%)

C:N

rat

io

Nut

rien

t (m

g/kg

)

N

P

K Ca

M

g S

Na

C

Cu

Zn

Mn

Fe

Al

B Si

Bios

olid

0.

81

0.39

1.

50

0.97

0.

77

0.64

2.

24

7.1

8.7

14.9

95

18

0 33

0 90

54

1.

38

MW

com

post

1.

85

0.28

0.

15

32.0

0.

19

0.13

0.

05

29.0

15

.7

56.8

12

0 83

67

0 22

00

11

15.8

6

Mill

Mud

1.

44

0.39

0.

54

6.30

0.

23

0.40

0.

52

19.8

13

.7

248.

0 46

0 27

0 13

000

1100

0 <

0.3

15.7

Mill

Ash

0.

30

0.27

0.

28

0.70

0.

16

0.05

0.

01

19.4

64

.7

32.4

76

93

0 13

000

3800

0 <

0.3

25.3

5

Bana

na t

rash

1.

82

0.35

2.

00

1.50

0.

72

0.10

0.

40

46.8

25

.7

59.3

11

0 13

00

2400

0 55

000

< 0.

3 2.

31

Gra

ss h

ay

2.28

0.

29

3.20

0.

49

0.16

0.

54

0.76

44

.7

19.6

9.

0 22

19

0 17

0 14

0 5.

9 1.

29

Alfa

lfa

hay

3.16

0.

25

1.70

1.

70

0.52

0.

35

0.30

45

.8

14.5

6.

1 15

61

88

0 70

0 45

0.

63

Mol

asse

s 0.

50

0.06

3.

80

0.72

0.

25

0.43

0.

04

- -

2 8.

8 70

35

0 11

0 2.

7 -


Table 3. Properties (mean ± standard error) of soil quality indicators and nem

atode comm

unities comparing banana cultivation to

undeveloped plant systems at four locations in north Q

ueensland.

Location East Palm

erston M

ission Beach Tully

Kennedy

Vegetation Banana

Pasture Banana

Organic

Rainforest Banana

Rainforest Pasture

Banana Rainforest

Indicators

Chem

ical

NO

3-N (kg/ha)

24±7 6±1

15±2 16±1

43±10 109±42

60±4 28±14

76±36 31±7

pH

6.5±0.2 5.4±0.1

5.2±0.1 5.5±0.3

4.6±0.1 5.5±0.3

4.3±0.2 4.1±0.1

4.3±0.3 5.2±0.1

EC soil (dS/m)

0.25±0.04 0.04±0.00

0.11±0.02 0.30±0.03

0.22±0.02 0.38±0.09

0.13±0.01 0.07±0.01

0.40±0.162 0.21±0.01

Physical

Temperature (C°)

24.6±0.1 26.1±0.1

20.9±0.2 20.9±0.1

20.4±0.1 18.9±0.1

18.5±0.3 20.2±0.1

19.4±0.2 19.6±0.0

Infiltration rate (s) 11±4

161±16 82±13

65±6 49±20

791±359 1195±320

1800±0 122±23

39±13Soil w

ater (g/g) 0.39±0.02

0.34±0.01 0.25±0.01

0.32±0.01 0.29±0.01

0.30±0.010 0.33±0.01

0.42±0.01 0.15±0.02

0.14±0.01Bulk density (g/cm

3) 0.86±0.02

0.91±0.01 1.12±0.01

1.04±0.02 1.07±0.02

1.12±0.06 1.03±0.02

1.06±0.01 1.24±0.01

0.92±0.05Aggregate stability (%)

62±1 66±3

62±2 76±1

78±2 20±4

71±4 74±3

41±5 80±5

Biological

Soil Respn (kg CO2 .ha

-1.d-1)

7.3±0.7 12.6±1.13

5.1±0.47 4.5±0.21

6.1±0.56 7.1±1.00

8.7±0.39 15.2±1.60

5.9±0.37 5.1±0.32

Root weight (g/L)

11.3±3.2 21.0±3.5

9.0±1.1 16.0±3.3

19.8±5.7 9.4±4.9

11.3±1.1 35.1±5.3

13.6±5.1 26.8±2.8

Soil surface trash (g/m2)

40±16 79±8

1.3±1.0 83±27

47±15 159±37

21±2 --

135±29 154±14

N

ematode proportions

Plant parasitic (%) 82±5

44±4 91±3

28±7 18±1

55±6 57±3

19±4 74±9

14±5Bacterial feeding (%)

13±4 20±2

6±2 18±3

33±3 23±6

13±2 26±2

17±6 47±4

Fungal feeding (%) 2±1

11±2 1±0

15±4 9±1

3±1 7±1

8±2 4±1

11±2Predatory (%)

0±0 2±1

1±0 23±4

7±1 5±1

9±1 5±1

1±1 6±1

Plant associated (%) 0±0

12±2 0±0

2±1 14±1

7±1 7±1

30±7 3±2

10±1O

mnivorous (%)

3±2 10±1

2±1 15±4

20±2 6±3

7±2 12±3

2±1 12±2

N

ematode indices

Shannon-Weiner (H

’) 1.39±0.10

1.93±0.08 0.76±0.15

1.76±0.08 2.24±0.08

1.80±0.10 2.21±0.07

2.00±0.05 1.39±0.19

2.23±0.07D

ominance (λ)

0.33±0.03 0.23±0.03

0.10±0.08 0.17±0.01

0.13±0.01 0.20±0.03

0.09±0.02 0.18±0.02

0.35±0.04 0.15±0.02

Bacterial.(Bact+Fung) -1 0.78±0.08

0.50±0.06 0.68±0.06

0.37±0.05 0.53±0.03

0.69±0.08 0.46±0.05

0.56±0.04 0.70±0.07

0.67±0.03Structural index

50±10 62±5

73±9 84±5

85±1 62±15

79±4 77±6

43±13 68±3

Enrichment index

63±5 51±5

82±4 67±5

83±3 82±5

66±6 82±4

70±5 77±3

Channel index 9.1±2.6

38.3±4.2 6.3±3.1

28.8±7.6 9.4±1.5

8.1±4.0 30.1±9.1

10.9±2.3 15.6±4.5

10.4±1.5


The total weight of banana roots was made up of the different root diameter classes (Figure 3). However, roots of different diameters were affected by different soil pro-perties. Roots greater than 5 mm were negatively correlated with soil bulk density and the proportion of bacterial feeding nematodes, but were positively correlated with aggregate stability and the proportion of plant feeding nematodes in the soil (Figure 3). The weight of roots 1-5 mm was positively correlated with the structure index (deter-

Aggregatestability

Nitratenitrogen

Channelindex

Gravimetricwater contentElectrical

conductivity

Soil surfaceorganic matter

Omnivores(%)

Fungal feedingnematodes (%)

Soilrespiration

Waterinfiltration rate

Plant parasiticnematodes (%)

Bacterial feedingnematodes (%)

Nematodediversity

Ratio of bacterialto fungal feeding

nematodes

Omnivorousnematodes

Plant associatednematodes

Nitrate-N

Fungal feedingnematodes

Predatorynematodes

Electricalconductivity

Aggregatestability

Waterinfitration rate

Soil bulkdensity

Plant feedinnematodes

Root weight> 5 mm

Leafemergence rate

Sucker heightincreaseNematode

dominance

Nematodediversity

Bacterialfeeding

nematodes

Worms

Structuralindex

Root weight< 1 mm

Total rootweight (g/L)

Root weight1-5 mm

Figure 2. Significant (P<0.05) correlations

between soil properties and nematode commu-nity analysis comparing

banana cultivation with undeveloped

plant systems in north Queensland. Dashed

lines (---) depict a negative relationship,

solid lines (____) depict a positive relationship.

Figure 3. Significant (P<0.05) correlations between soil proper-

ties and nematode communities within 21 banana fields in north

Queensland. Dashed lines (---) depict a

negative relationship, solid lines (____) depict a positive relationship.


mined by the nematode community analyses as a measure of the trophic layers and potential for opportunists, Ferris et al. 2001), but negatively correlated with nematode dominance and water infiltration rates (Figure 3). Roots less than 1 mm in diameter were positively correlated with electrical conductivity (Figure 4). Aggregate stability was again negatively correlated with nitrate-N amounts in the soil (Figure 4). There was

Root weight< 1 mm

Root weight1-5 mm

Root weight> 5 mm

FingerNumber

Soilrespiration

Soiltemperature

Sucker heightincrease

Soil pH

Worms

Soil bulkdensity

Total rootweight (g/L)

Enrichmentindex

Channelindex

Structuralindex

Nematodediversity

Nematodedominance

Leafemergence

rateNitrate-N

Electricalconductivity

Omnivorousnematodes

Fungal feedingnematodes

Bacterialfeeding

nematodes

Plantassociated

Plant-parasiticnematodes

Predatorynematodes

Ratio of bacterialto fungal feeding

nematodes

Figure 4. Significant (P<0.05) correlations between soil properties and nematode communities within a banana field in north Queensland sampled monthly 9 times since October 2002. Dashed lines (---) depict a negative relationship, solid lines (____) depict a positive relationship.

no correlation between soil factors and plant production parameters (Figure 4), which suggested that farmer management of the crops may be able to obscure poor soil and root characteristics.

Banana field surveyIn a single banana field there was an increasing number of relationships between soil properties (Figure 4). Plant parasitic nematodes were not the dominant trophic nema-tode group at this site (Table 5). The structure index (Ferris et al. 2001) was positively correlated with the total weight of roots and the weight of roots between 1-5 mm (Figure 4). The channel index (33) was almost double the banana industry average (19), which suggested that fungi were more important in the recycling of nutrients even though bacterial feeding nematodes dominated the nematode community (Table 5). The site also had low soil nitrate-N (14 kg NO3-N.ha-1), approximately half of the north


Queensland industry mean (33 kg NO3-N.ha-1), which meant that recycling of nutrients is more important to supply nitrogen for crop production at this site (Tables 4 and 5).

The site was measured over different environmental conditions and soil temperature was positively correlated with the increase in follower sucker height and soil pH and negatively correlated with bunch finger number (Figure 4). This suggested that seaso-nal fluctuations may impact more on crop production than on soil characteristics and obscure effects of soil health on crop production.

Table 4. Properties (grand mean ± standard error) of soil quality indicators and nematode communities of 21 banana fields in north Queensland, n=105.Soil property Mean ±Standard error

Physical properties

Soil temperature (°C) 24.2 0.5Water infiltration rate (secs) 448 108Gravimetric soil water (g/g) 0.25 0.02Soil bulk density (g/cm3) 1.11 0.03Aggregate stability (%) 43.9 4.2

Chemical properties

Electrical conductivity (dS/m) 0.25 0.04pH 6.13 0.16No3-N (kg/ha) 32.9 6.6

Biological properties

Soil respiration (kg CO2-C ha-1 d-1) 6.53 0.30Worms (0=absent, 1=present) 0.4 0.1

Banana measurements

Finger number per bunch 133 9Leaf emergence rate (leaves/week) 0.50 0.04Sucker height increase (m/week) 0.06 0.01Total root weight (g/L) 8.19 0.83Root weight >5 mm (g/L) 3.19 0.40Root weight 1-5 mm (g/L) 2.45 0.24Root weight <1 mm (g/L) 2.56 0.41Surface trash (t/ha) 8.1 1.1

Soil nematodes

Plant parasitic nematodes (%) 61.7 4.8Bacterial feeding nematodes (%) 21.5 2.5Fungal feeding nematodes (%) 4.5 1.1Predatory nematodes (%) 4.2 0.9Plant associated nematodes (%) 2.0 0.9Omnivorous nematodes (%) 6.1 1.1

Nematode indices

Shannon-Weiner diversity index H′ 1.11 0.09Dominance λ 0.37 0.04Bacterial.(Bacterial+Fungal)-1 0.82 0.03Structural index 60.3 3.1Enrichment index 68.3 2.6Channel index 18.6 4.3


Soil amendmentsThe addition of mill mud (40 tonnes/ha) or mill ash (120 tonnes/ha) significantly increased the weight of banana roots relative to the untreated control (Table 6). However, only the addition of mill mud was able to significantly increase the shoot weight (Table 6). The addition of biosolids to the soil significantly reduced the weight of banana roots and shoots (Table 6).

R. similis numbers within the roots of bananas were significantly suppressed when grass hay, alfalfa hay, mill mud and banana trash were added to the soil (Table 6).

Table 5. Properties (grand mean ± standard error) of soil quality indicators and nema-tode communities within a banana field in north Queensland sampled 9 times at monthly intervals from October 2002.Soil property Mean ±Standard error

Physical properties

Soil temperature (°C) 25.5 0.8Water infiltration rate (secs) 350 129Gravimetric soil water (g/g) 0.27 0.01Soil bulk density (g/cm3) 1.13 0.01Aggregate stability (%) 47.6 2.4

Chemical properties

Electrical conductivity (dS/m) 0.18 0.02pH 6.47 0.29NO3-N (kg/ha) 13.6 1.9

Biological properties

Soil respiration (CO2-C ha-1 d-1) 5.6 0.4Worms (0=absent, 1=present) 0.4 0.1Banana measurementsFinger number per bunch 101 6Leaf emergence rate (leaves/week) 0.7 0.1Sucker height increase (m/week) 0.07 0.02Total root weight (g/L) 6.59 0.89Root weight >5mm (g/L) 1.77 0.47Root weight 1-5mm (g/L) 2.53 0.38Root weight <1mm (g/L) 2.29 0.36Surface trash (t/ha) 5.2 0.7

Soil nematodes

Plant parasitic nematodes (%) 23.6 2.7Bacterial feeding nematodes (%) 37.0 4.6Fungal feeding nematodes (%) 19.2 3.6Predatory nematodes (%) 7.0 1.8Plant associated nematodes (%) 1.8 0.8Omnivorous nematodes (%) 11.3 2.1

Nematode indices

Shannon-Weiner diversity index H′ 1.56 0.09Dominance λ 0.24 0.02Bacterial.(Bacterial+Fungal)-1 0.65 0.07Structural index 67.7 5.5Enrichment index 61.8 4.3Channel index 33.1 7.7Means are calculated from 45 samples.


There was no interaction between soil type and the amendments, which suggested that the effects of the amendments would be the same on banana growth and R. similis in banana producing soils of north Queensland.

DiscussionThe health of the banana root system is linked to the conditions of the soil in which the plants grow. Complex interactions between soil physical, chemical and biological properties can be mapped to gain a greater understanding of the interactions of crop production and soil health under specific soil conditions. As the cropping systems and conditions in which the plants are grown become more uniform, the interactions within the soil become more distinct. This was seen in moving from a survey of different plant systems, to a survey of banana growing systems finally to a single banana field (Figures 2, 3 and 4).

Skilful farm management can mask the deterioration of the soil and obscure interactions of soil components to obtain profitable banana production. It is not until the plant is pla-ced under stress that soil degradation becomes apparent within the banana production system. The time needed for the soil to respond and overcome an imposed stress may be one method of determining the health of the soil (van Bruggen and Semenov 2000). Damage to banana roots by R. similis can be interpreted as a stress on the soil ecosys-tem. The nematode is capable of significantly reducing the root system of banana plants even in favourable physical soil conditions (Figure 1). Therefore, improved biological and chemical management of the soil appear as important in suppressing nematode damage to banana roots as developing an extensive root system.

Tools and indicators are needed to measure changes in soil properties to pre-vent soil degradation. The basic soil quality indicators (Sarrantonio et al. 1996, http://soils.usda.gov/sqi/soil_quality/assessment/kit2.html) used in the BRASH project

Table 6. Effects of applications of soil applied amendments on R. similis numbers in the roots of bananas and the weight of banana roots and shoots. Amendment Rate Nematodes Shoot Root (t/ha) in 100 g root dry weight (g) fresh weight (g)

Untreated - 10 046 d 17.8 bcd 53.8 bc

Bedminster 40 7 449 d 18.1 bcd 54.8 bcd

Biosolid 40 3 935 bcd 11.6 a 36.0 a

Grass 40 1 734 b 21.2 cde 61.4 bcde

High ash 120 3 931 bcd 22.0 de 71.4 de

Low ash 40 5 452 cd 21.1 cde 52.9 abc

Lucerne 40 569 a 15.7 ab 50.5 abc

Mill mud 40 2 121 bc 25.9 e 75.9 e

Molasses 300 L/ha 8 847 d 16.6 bc 46.1 ab

Paunch 40 4 354 bcd 17.5 bcd 45.6 ab

Trash 40 1 778 b 24.5 e 66.9 cde

Wolastonite 5 9 661 d 16.8 bc 56.4 bcd

Means with the same letter within a column are not significantly different at the 5% level.


showed relationships with nematode indicators (Yeates and Bongers 1999, Ferris et al. 2001). This suggested that simple tools can indicate more complex soil processes. The tools can be used to depict how farmer’s soil management decisions can have ‘flow on’ effects that are not easily recognisable. In the survey of banana farms in north Queensland the accumulation of soil nitrate was positively correlated with EC and negatively correlated with aggregate stability (Figures 2 and 3). Aggregate stability in turn was negatively correlated with water infiltration rates, which were negatively correlated with the total weight of banana roots (Figure 3). As EC increased so did the weight of roots less than 1 mm diameter. It could be proposed that as EC increased the banana plant produced more fine roots, possibly to absorb the readily available nitrate-N. Therefore, the use of nitrate-N fertilisers in the short term may produce higher yields, however, the long-term effects may lead to soil degradation as aggregates become less stable. Excessive soil nitrogen has also been linked with increased soil-borne diseases in crops due to the lack of microbial diversity (van Bruggen and Termorshuizen 2003). This requires further investigation in banana production.

Abawi and Widmer (2000) suggested that soil-borne diseases are most damaging when soil conditions are poor as a result of inadequate drainage, poor soil structure, low orga-nic matter, low soil fertility and high soil compaction. However, effects of soil health on the suppression of soil-borne diseases need to be extrapolated to the banana field. Management of the soil microbial environment may be able to reduce the susceptibility of bananas to soil-borne diseases. Increasing the nematode diversity in the soil was negatively correlated with the proportion of plant-parasitic nematodes (Figures 2, 3 and 4). The addition of soil amendments high in carbon was found to suppress R. similis in a pot experiment (Tables 1 and 6). Similarly, Stirling et al. (2003) found that amendments with high C:N ratios are the most effective in enhancing biological control activity against plant-parasitic nematodes. One possible mechanism for the suppression of plant parasitic nematodes is that there are a variety of nematode parasitic fungi in the soil that use nematodes as a source of nitrogen (Stirling et al. 2003). Conditions in the field need to be manipulated to increase soil organisms that suppress plant-parasitic nematodes. This has been promoted as longer control than the use of chemicals (Widmer et al. 2002) and may promote a more efficient banana root system.

One important soil characteristic that is not easily measured is soil organic carbon. Widmer et al. (2002) suggested that maintenance of high concentrations of organic mat-ter, especially the active fraction, greatly improves the physical, chemical and biologi-cal properties of soils leading to increased productivity. Tropical soils used in banana production tend to have high soil water contents and high soil temperatures, which are favourable for organic matter decomposition (Sikora and Stott 1996). Additionally, intensive cultivation of the soil in preparation for planting bananas in north Queensland may also be reducing soil carbon. A simple on farm test to determine soil organic car-bon is needed to allow monitoring and linking in to other soil health indicators.

The commercial production of bananas in monoculture, using the same clone, makes commercial bananas susceptible to pests and diseases due to the lack of diversity. Yeates (1999) found that plant communities were responsible for dictating soil nematode com-munities. The use of ground-cover plants may be one method of increasing the diversity around the banana root system and improving soil physical, chemical and biological


properties. The mechanized method of banana cultivation in north Queensland would allow mowing of ground-cover plants to return organic matter to the banana soil eco-system. The choice of species is important and ideally would be deep rooted, allowing the capture of nutrients that escape the banana root system, shade tolerant and capable of withstanding waterlogging and traffic.

The use of soil health indicators needs to be linked with a framework of continual crop and environmental improvement, to support management decisions if farmers are to make advances in environmentally responsible soil management. However, soil mana-gement issues may be specific to individual farming systems. Heisswolf et al. (2003) designed a process that valued farmer’s motives, knowledge and experience, was par-ticipative and encouraged farmers to identify their own environmental issues within an economic perspective. This process could be linked with the tools and indicators used in monitoring soil health and allow farmers to understand the effects their management is having on soil health. However, soil management must be seen as a holistic system, made up of physical, chemical and biological components that cannot be separated.

The health of the banana root system is strongly reliant on soil health determined by soil physical, chemical and biological properties. The soil properties are linked within the soil ecosystem, so that subtle management changes can have long-term effects leading to soil degradation and future problems with banana production. The BRASH project aims to involve farmers in changing soil management practices by demonstra-ting the impact that soil degradation, such as soil compaction, has in reducing banana root development. BRASH develops simple tools for farmers to monitor the trends in soil health and provides solutions to correct adverse trends, such as the use of soil amendments. However, due to complex interactions occurring in the soil and the long time frame needed to detect changes in soil indicators, a framework for integrating and managing all of the components is required if sustained changes in environmental management practices are to occur. Most banana production world-wide is in environ-mentally sensitive of areas. Therefore, responsible soil management is needed to ensure that banana production remains socially and economically viable with minimal impact on the surrounding environment.

AcknowledgementsHorticulture Australia Ltd and Queensland Fruit and Vegetable Growers, Banana Committee provided funding for this work under project FR02025. Tanya Martin, Mark Poljak and Tony Nikulin provided technical assistance for the research involved. The contributions and time of the banana growers of north Queensland towards this project is greatly appreciated, in particular G. Bush, M. Franklin, P. Leahy and D. Wall.

ReferencesAbawi G.S. & T.L. Widmer. 2000. Impact of soil health management practices on soilborne pathogens, nematodes and

root diseases of vegetable crops. Applied Soil Ecology 15:37-47.

Araya M. & F. Blanco. 2001. Changes in the stratification and spatial distribution of banana (Musa AAA cv. Grand Naine) root system of poor, regular and good developed plants. Journal of Plant Nutrition 24:1679-1693.

Blake C.D. 1969. Nematode parasites of banana and their control. Pp. 108-132 in Nematodes of Tropical Crops. (J.E. Peachey, ed.). Commonwealth Bureau of Helminthology, St. Albans.


Bongers T. 1990. The maturity index: an ecological measure of environmental disturbance based on nematode species composition. Oecologia 83: 14-19.

Delvaux B. 1995. Soils. Pp. 230-257 in Bananas and Plantains. (S. R. Gowen, ed.). Chapman and Hall, London.

Doran J.W. 2002. Soil health and global sustainability: translating science into practice. Agriculture, Ecosystems and Environment 88:119-127.

Doran J.W. & T.B. Parkin. 1996. Quantitative indicators of soil quality: A minimum data set. Pp. 25-37 in Methods for Assessing Soil Quality. (J. W. Doran & A. J. Jones, eds). Soil Science Society of America, Inc., Madison.

Ferris H., T. Bongers & R.G.M. de Goede. 2001. A framework for soil food web diagnostics: extension of the nematode faunal analysis concept. Applied Soil Ecology 18:13-29.

Fogain R. 2001. Nematodes and weevil of bananas and plantains in Cameroon: occurrence, importance and host suscep-tibility. International Journal of Pest Management 47:201-205.

Freebairn D.M. & C.A. King. 2003. Reflections on collectively working toward sustainability: indicators for indicators! Australian Journal of Experimental Agriculture 43:233-238.

Gowen S.R. & P. Quénéhervé. 1990. Nematode parasites of bananas, plantains and abaca. Pp. 431-460 in Plant parasitic nematodes in subtropical and tropical agriculture. (M. Luc, R. A. Sikora, & J. Bridge, eds). CAB International, Wallingford.

Gowen S.R. 1995. Pests. Pp. 382-402 in Bananas and Plantains. (S.R. Gowen, ed.). Chapman and Hall, London.

Heisswolf S., S. Lindsay, J. Bagshaw & N. Vock. 2003. New experiences in working with horticultural farmers to impro-ve NRM practices in Queensland. 2003 APEN national forum 26-29 November 2003, Hobart, Tasmania, Australia, Australasian Pacific Extension Network: http://www.regional.org.au/au/apen/2003/2/110heisswolfs.htm

Hooper D.J. 1986. Extraction of nematodes from plant material. Pp. 51-58 in Laboratory methods for work with plant and soil nematodes. (J. F. Southey, ed.). Her Majesty’s Stationary Office, London.

Lobry de Bruyn L.A. & L.A. Abbey. 2003. Characterisation of farmers’ soil sense and the implications for on-farm monitoring of soil sense. Australian Journal of Experimental Agriculture 43:285-305.

Marsh S.P. 1998. What can agricultural researchers do to encourage the adoption of sustainable farming systems. Sustainability of Farming Systems with Focus on Future Management of Water and Nutrients, Perth, Western Australia.

Moens T.A.S., M. Araya & D. DeWeale. 2001. Correlation between nematode numbers and damage to banana (Musa AAA) roots under commercial conditions. Nematropica 31:55-65.

Neher D.A. 2001. Role of nematodes in soil health and their use as indicators. Journal of Nematology 33:161-168.

Price N.S. 1995. Banana morphology - part 1: roots and rhizomes. Pp. 179-189 in Bananas and Plantains. (S. R. Gowen, ed.). Chapman and Hall, London.

Sarrantonio M., J.W. Doran, M.A. Liebig & J.J. Halvorson. 1996. On-farm assessment of soil quality and health. Pp. 83-105 in Methods for Assessing Soil Quality. (J.W. Doran & A.J. Jones, eds). Soil Science Society of America, Madison.

Schloter M., O. Dilly & J.C. Munch. 2003. Indicators for evaluating soil quality. Agriculture, Ecosystems and Environment 98:255-262.

Sikora L.J. & D.E. Stott. 1996. Soil organic carbon and nitrogen. Pp. 157-167 in Methods of Assessing Soil Quality. (J.W. Doran & A.J. Jones, eds). Soil Science Society of America, Madison.

Stamatiadis S., M. Werner & M. Buchanan. 1999. Field assessment of soil quality as affected by compost and fertiliser application in a broccoli field (San Benito County, California). Applied Soil Ecology 12:217-225.

Stirling G.R., E. Wilson, M. Stirling, C.E. Pankhurst & P. Moody. 2003. Organic amendments enhance biological suppression of plant-parasitic nematodes in sugarcane soils. Proceedings of the Australian Society of Sugarcane Technologists: 13.

Turner D.W., J.C. Mulder & J.W. Daniells. 1988. Fruit numbers on bunches of bananas can be estimated rapidly. Scientia Horticulturae 34:265-274.

van Bruggen A.H.C. & A.V. Semenov. 2000. In search of biological indicators for soil health and disease suppression. Applied Soil Ecology 15:13-24.

van Bruggen A.H.C. & A.J. Termorshuizen. 2003. Integrated approaches to root disease management in organic farming systems. Australasian Plant Pathology 32:141-156.


Whitehead A.G. & J.R. Hemming. 1965. A comparison of some quantitative methods extracting small vermiform nema-todes from the soil. Annals of Applied Biology 55:25-38.

Widmer T.L., N.A. Mitkowski & G.W. Abawi. 2002. Soil organic matter and management of plant-parasitic nematodes. Journal of Nematology 34:289-295.

Yeates G.W. 1979. Soil Nematodes in Terrestrial Ecosystems. Journal of Nematology. 2:213-229.

Yeates G.W. 1999. Effects of plants on nematode community structure. Annual Review of Phytopathology 37:127-149.

Yeates G.W. 2001. Diversity of soil nematodes as an indicator of sustainability of agricultural management. Australasian Plant Pathology Society, Nematology Workshop, Cairns, Queensland, Australia.

Yeates G.W. & T. Bongers. 1999. Nematode diversity in agroecosystems. Agriculture, Ecosystems and Environment 74:113-135.


5

Pathogen - root interactions

Interacciones patógeno - sistema radical

C.A. Gauggel et al. 169A. Zum Felde et al.

The potential use of microbial communities inside suppres-sive banana plants for banana root protectionAlexandra zum Felde1*, Luis Pocasangre2 and Richard A. Sikora1

AbstractResearch has demonstrated that plants lose up to 33% of their assimilates to the soil. Why does a plant, in this case banana, exert large amounts of energy to produce nutrients that land unused in the soil? We believe that: (1) roots are damaged by pests and diseases and the plant cannot utilize the nutrients produced in the shoot; and/or (2) the plant has evolved a health support system made up of “rhizosphere specific microbial communities” (RSMC) that live on these nutrients in symbiotic and/or mutualistic associations. These RSMC are not enhanced by standard banana production systems even though they are important for root health and growth. Research has shown that the interactions between banana and certain endophytic fungi are important for root health and growth. This interaction has been studied in detail and these forms of microbial communities have evolved concomitantly with the plant over evolutionary time. We have shown that specific fungi, and probably even bacteria, that have plant health promoting abilities are important for root health. When an ecological state is reached in which RSMC are well established and are functioning properly, we believe this leads to a disease or pest suppressive agro-ecosystem. Such a system was studied in Guatemala. The research indicated that: 1) pest nematodes are suppressed in certain areas of the Motagua Valley, Guatemala; 2) endophytic fungi play a major role in this suppressive system, and 3) endophytic Fusarium oxysporum and Trichoderma atroviride suppress Radopholus similis in banana roots.

Resumen - Uso potencial de las comunidades microbianas en plantas supresivas de banano para proteger las raíces del banano

Investigaciones han demostrado que las plantas pierden hasta un 33% de sus asimilados en el suelo. ¿Porqué una planta, en este caso banano, utiliza grandes cantidades de energía para producir nutrientes que se pierden en el suelo? Nosotros creemos que: (1) las raíces son dañadas por plagas y enfermedades y no pueden utilizar los nutrientes producidos en en la parte aérea de la planta; y/o (2) la planta ha desarrollado un sistema de apoyo a la salud con “comunidades microbiales específicas de la rizosfera” (CMER) que viven en asociaciones simbióticas y/o mutualísticas con la planta y que se alimentan con esos nutrientes. Las CMER no se mejoran mediante los sistemas de producción estándar de banano, aún cuando éstas son importantes para el crecimiento y la salud de la raíz. La investigación ha mostrado que la interacción entre el banano y hongos endofíticos específicos son importantes para la salud y el crecimiento de la raíz. Esta interacción ha sido estudiada en detalle y estas formas de comunidades microbiales se han desarrollado en concomitancia con la planta sobre un tiempo evolucionario. Hemos demostrado que hongos específicos y probablemente aún bacterias que tienen la habilidad de promover la salud de la planta, son importantes para la salud de las raíces. Creemos que cuando se alcanza un estado ecológico en el cual las CMER están bien establecidas y funcionando apropiadamente, da como resultado un agro-ecosistema supresivo de enfermedades o plagas. Un sistema de este tipo fue estudiado en Guatemala. La investigación indicó que: 1) nematodos parásitos son suprimidos en algunas sitios en el Valle Motagua en Guatemala; 2) hongos endofíticos son importantes en este sistema supresivo, y 3) aislados endofíticos de Fusarium oxysporum y de Trichoderma atroviride suprimen Radopholus similis en las raíces del banano.

1 Soil Ecosystem Phytopathology and Nematology, Institute for Plant Disease, University of Bonn, Nussallee 9, 53115 Bonn, Germany; 2 INIBAP-LAC/CATIE, Turrialba, Costa Rica. *Corresponding author: e-mail: [email protected]

The problem of banana root deterioration and its impact on production170 The potential use of microbial communities inside suppressive banana plants... C.A. Gauggel et al. 171A. Zum Felde et al.

IntroductionThe idea that susceptible plants can survive in a field heavily infested with an econo-mically important pest or disease because they have acquired high levels of microbial-based suppression is fascinating, but not impossible. The banana is a perennial plant that, under commercial and subsistence production remains in a soil for up to 25 years. This exerts selection pressure not only on pests and diseases but also on beneficial microorganisms in the soil surrounding the root. Under banana growing conditions, the selection of these important groups of organisms out of the total soil biodiversity is a 24 hour, 365 day and multiple year process that leads either to more disease or more antagonistic potential (Sikora 1992, 2002).

The presence of suppressive soils is well known in other crops. In banana it has been overlooked due to the devastating effect of the burrowing nematode, Radopholus similis, on banana root health. It has also gone unnoticed because of the multiple and yearly applications of nematicides on banana plantations. This masks the existence of pest suppression and ultimately the detection of this important phenomenon. In other cropping systems, it has been shown that monoculture can lead to the development of high levels of antagonists in the soil and finally to suppression, or a virtual elimination of a pest nematode from the soil (Kerry et al. 1982). Rhizosphere and endorhiza specific microbial communities are selected by root exudates (Curl and Truelove 1986, Vilich and Sikora 1998). However, this does not mean that every plant in a field will have the exact same microbial community, as microorganisms are not equally distributed. This is another reason why plants suppressive to a pest have been overlooked.

If we hope to reduce the use of nematicides in banana production, we either need a plant with durable resistance or we need to increase the plant’s defences against that pest. Resistance is without doubt a priority, but increasingly aggressive races of R. similis are likely to develop under the constant pressures of monoculture. Therefore, additional tools are necessary to sustain yield, protect plant resistance, lower pesticide use and reduce overall costs. We believe a biological system management strategy (Sikora 1997) that integrates an understanding of the cropping system, the pest and the biological factors that can limit infection is the right approach.

Suppression has been equated with “The Edge of Chaos”, which is the dynamics of an ecosystem in which individual microbes optimize themselves until they reach sustained fitness (Sikora and Reimann 2004). In our opinion, “The Edge of Chaos” is the norm and not the exception in a banana plantation. However, it does not exist on every plant or it would have been detected long ago. Constant and prolonged selection of rhizos-phere specific microbial communities is the rule in and around a banana plant. As stated above, all plants secrete certain exudates that select specific microbial communities in the soil. Therefore, root exudates specific to a banana cultivar select a rhizosphere- and endorhiza-specific microbial community, which is involved in susceptibility, tolerance and/or resistance to pest infection. It must be made clear that these communities can be detrimental or beneficial to the plant with a balance between the two being the most common case. The question is: can we manipulate this state to the benefit of the banana plant?


Questions that banana producers need to ask themselves are:

1. Does suppressiveness exist in or on a banana plant or in banana plantations?

2. Can plant-beneficial nematode-antagonistic microorganisms, living in the rhizosphere or endorhiza, be isolated from the microbial community?

3. Is it possible to detect specific microbes in banana that are highly effective in reducing or suppressing nematodes in banana roots?

4.: Can some form of suppression be established in banana using these beneficial microbes?

5. Can full suppression be achieved either by using a single microorganism or by a multi-trophic inundative system?

6. Is enhancement of banana planting material an effective means of attaining biological system management in the field?

We believe that the answer to all six questions is a firm yes! We believe this response to be valid based on past experience in banana research. This has been demonstrated in research results obtained in over 15 years of work on the antagonistic potential of microorganisms for nematode, insect and disease control with banana in cooperative research with ICIPE, IITA and INIBAP. The concepts, methods and results that cover the development of this biological system management approach will not be reviewed here but have been published elsewhere (Sikora and Schuster 1998). As part of the ongoing research to detect suppressive plants and associated microorganisms, a survey of endophytic fungi from banana and the screening of their nematode-antagonistic potential was carried out in Central America in 1998 (Niere et al. 1998, Pocasangre et al. 2000). Following this study, reports of nematode suppressive soils in Guatemala focused our attention on the Motagua Valley. Such areas, where natural suppression of pathogens occurs, represent an ideal source of potential bio-control organisms (Dicklow et al. 1993). To study this area, a research programme was established in Guatemala and Costa Rica with the cooperation of INIBAP, CATIE and BANDEGUA, during the 2001-2002 growing season. The research programme had the following goals:

1. To verify if the three main plant parasitic nematodes affecting banana are sup-pressed in the banana production districts of the Motagua Valley, Guatemala;

2. To isolate and identify the endophytic fungi present in the root tissue of bana-nas from these districts;

3. To screen endophytes in vivo for antagonistic effects toward R. similis.

Materials and methodsSampling took place in randomly selected plots on one farm per banana growing dis-trict in the Motagua Valley of Guatemala; namely on El Real farm in the Panchoy dis-trict, Maya in Motagua A, Creek in Motagua B, and Lourdes in Bobos. Material came only from ‘Grande naine’ (AAA) banana plants that had flowered within 14 days of sampling. Sampling was done according to the method described by Speijer and Gold (1996) and restricted to the area between the base of a mother plant and its follower


sucker. Composite root samples were made up of the excavated roots of ten plants encountered in a plot.

Sampling took place on two days, farms in the Motagua A and Motagua B districts were sampled on 22 October 2001, and farms in the Bobos and Panchoy districts were sampled on 26 October 2001. Root necrosis was assessed by washing the total fresh root sample with tap water, weighing it and dividing the roots into necrotic and healthy roots. The necrotic roots alone were then weighed and the percent of necrotic roots (%NR) calculated using the following formula:

%NR = (fresh weight of necrotic roots/fresh weight of total roots) x 100

As composite samples of ten plants were collected, the weight of functional roots per plant was calculated by dividing the fresh weight of functional roots per composite sample by ten.

The method used to extract plant-parasitic nematodes from the roots was based on the maceration-sieving method described by Speijer and De Waele (1997). To assess the nematode densities in the roots, a 25 g sub-sample of healthy root tissue was remo-ved from each sample. Roots were chopped and macerated in a kitchen blender with approximately 200 ml tap water for 12 s at low speed. The sample was then transferred to four nested screen filters with apertures of 600 μm, 150 μm and two of 45 μm, filte-red and washed with tap water. The contents of the final two screens were poured into a clean 1000 ml-graduated cylinder that was filled with tap water. The contents of the cylinder were shaken well and 2 ml removed for counting using a nematode counting slide. All nematodes present in the sub-sample were identified and counted, and the number of nematodes per 100 g root estimated.

To determine and compare the endophytic fungal populations of the four production areas, endophytic fungi were isolated from roots by placing surface sterilized root pieces on selective growth media. Three to 5 cm-long healthy root segments were sterilized in 1.5% NaOCl solution for 5 minutes, washed with sterile water and dried on sterile filter paper. The ends of each root segment were trimmed and the segment cut lengthwise. Five such lengthwise cut pieces, each from a different root segment, were plated on 90 mm Petri plates with 10% potato dextrose agar (PDA) plus antibiotics (4 g PDA, 16 g agar, 150 mg Penicillin G and 75 mg Streptomycin Sulphate per litre distilled water). Plates were stored in a dark incubator at 25°C and regularly checked for fungal growth. All fungal colonies encountered over a six-week period were described, pure cultures produced on fresh 10% and 100% PDA plates, and later identified. A selection of the endophytes recovered was screened for antagonistic effects towards R. similis.

Twelve endophyte isolates were selected for screening of antagonistic activity towards R. similis on micropropagated banana plantlets of the cultivar ‘Grande naine’. Micro-propagated plantlets were obtained from a commercial biotech company in San Jose, Costa Rica. The test was set up in a totally randomized block design with seven repli-cations of 14 treatments: 12 fungal isolates and 2 controls. The controls consisted of an entry without fungi, but inoculated with R. similis (Control); and the absolute control, free of fungi and nematodes (AC). Since the effect of the fungi on nematode density in plant roots was the primary objective of this test, no controls with plants inoculated exclusively with fungi were used.


The selected endophytic Fusarium and Trichoderma isolates were grown on 100% PDA plates until they produced sufficient conidia: about 8 days in the case of Fusarium spp., and 3-5 days in the case of Trichoderma spp. Under a laminar-flow hood, conidia were removed from plates by pouring sterile, distilled water onto the agar surface and gently moving the water over the surface with a cool, flame-sterilized glass scraper. The resulting conidia stock solution was poured through sterile gauze into sterile 50 ml Erlenmeyer flasks and capped with aluminium foil. Conidia concentrations of the stock solutions of each isolate were determined with the help of a Neubauer haemocytometer. For each fungal isolate screened, 300 ml of inoculation suspension with 1 x 105 conidia/ml were prepared with calculated volumes of distilled water and stock solution, in 500-ml beakers. In the greenhouse, the roots of micro-propagated ‘Grande naine’ plantlets ~12 cm tall were immersed and gently shaken in the inoculation solution for 5 minutes, and immediately planted in 1 L pots containing a 1:1 mix of sterile river sand and soil. The roots of the two sets of controls were immersed in distilled water for 5 minutes. All plants were left to harden under greenhouse conditions for two weeks (tempera-ture: 25oC ± 2oC, natural light conditions, daily irrigation with tap water, no fertilizer). They were then inoculated with 500 R. similis nematodes extracted from carrot disks, provided by T. Moens from CORBANA. After eight weeks, the plants were removed from pots and shoot and root fresh weight recorded. Nematodes were extracted from roots according to the method described by Speijer and De Waele (1997). Nematodes in two 3 ml aliquots per plant were counted and the total number of nematodes per 100 g root was calculated.

ResultsThe total number of nematodes recovered per 100 g root differed significantly from farm to farm (Figure 1), and supported the designation of El Real and Maya farms as suppressive, of Creek farm as moderately suppressive and of Lourdes farm as

Figure 1. Total number of nematodes recovered from banana roots from sampled farms in October 2001.Columns with different letters are significantly different at P ≤ 0.05, using the one-way ANOVA test in StatsGraphics Plus 3.1.

Soil density (g/cm3)

c

b

a

a

0

10000

20000

30000

40000

50000

60000

70000

80000

El Real Maya Creek LourdesFarms

Meloidogyne spp.

Helicotylenchus spp.

Radopholus similis

#ne

matod

es/10

0groot


non-suppressive. Meloidogyne spp. populations neither differed significantly between farms, nor did their size indicate an important infestation (Table 1). The same can be said for the Helicotylenchus spp. populations, though their numbers were significantly greater on the Lourdes farm. However no suppression was expected nor observed in this area, so the results support the assumption that suppression is only present in cer-tain areas of the Motagua Valley. Root health indices verify that the roots of the sup-pressive plants from El Real and Maya farms are larger and significantly less necrotic than those from Creek or Lourdes farms.

Table 1. Number of nematodes recovered from banana roots from sampled farms in October 20011 and root health indices.

Origin Number of nematodes recovered per 100 g root3 Root health indices

(Farm)2 Radopholus Helicotylenchus Meloidogyne Weight of % necrotic similis spp. spp. functional root (g) root

El Real 400a 10 500a 2 800a 90b 3a

Maya 800a 3 200a 600a 132c 4a

Creek 21 800b 12 600a 1 000a 72ab 12b

Lourdes 51 400c 22 900b 500a 69a 22c1 Numbers followed by the same letter in a column are not different according to the one-way ANOVA test in StatsGraphics Plus 3.01 (P ≤ 0.05). 2 Farms are presented in order of suppressiveness: El Real – suppressive (SS); Maya – suppressive (SS); Creek – modera-tely-suppressive (MS); and Lourdes – non-suppressive (NS).3 All life stages of nematodes were included in the count and only functional roots were used for nematode extraction.

Table 2. Genus and number of fungi and their proportion of the population isolated from banana roots from four farms.

Genus of fungal Sampled farms and their suppressive status

isolate El Real – SS1 Maya - SS Creek - MS Lourdes - NS

(%) (%) (%) (%)

Fusarium 2 5 22 41 54 61 53 69Trichoderma 2 5 11 20 7 8 4 5Penicillium 2 5 4 7 3 3 4 5No ID or Lost2 8 20 10 19 24 27 16 21

Arthrobotrus 1 3 - -Aspergills - - 1 2 Basiptospora - - 1 2 Candida 4 10 - - Chrysosporium 12 30 - - Gleosporium 1 3 2 4 Gliocephalis 1 3 - - Gliodadum 1 3 - - Gliodium 1 3 - - Paecilomyces - - 1 - Torula / Wallemia - - 1 - Trichotechium 3 8 1 2 Tritirachium 1 3 - - Verticillium 1 3 - - Total 40 (100) 54 (100) 88 (100) 77 (100)1 SS: Suppressive; MS: Moderately suppressive; NS: Not suppressive. 2 Isolates lost to bacterial contamination were con-sidered lost and not identified.

Further identification of isolates from Creek and Lourdes was not carried out, as no nematode suppression was observed in these farms


DiscussionThis latest study confirms that nematode suppression exists in banana plants and in banana plantations. Nematodes are strongly suppressed in the Panchoy and Motagua A districts in the Motagua Valley. Additionally, we managed to isolate a variety of plant-beneficial nematode-antagonistic microorganisms living in the endorhiza. Among

Table 3. Screening of 12 fungal endophyte isolates for antagonistic activity towards Radopholus similis on micropropagated plantlets of the cultivar ‘Grande naine’ (AAA)1. Isolate code Number of Reduction Root weight Shoot weight R. similis2 % g3 g4

Control5 18 694 a - 8.3 b 20.0 bc

MF-50 14 707 a -21 10.8 ab 17.9 c

MF-22 14 025 ab -25 11.3 ab 19.9 bc

MF-30 12 931 ab -31 13.3 ab 19.1 bc

MF-10 11 171 abc -40 10.3 ab 22.7 abc

EF-39 7 890 bdc -58 14.5 a 28.3 a

MF-49 7 417 dc -60 14.2 a 26.3 abc

MF-6 6 764 dc -64 10.3 ab 25.4 ab

MF-12 6 601 dc -65 12.4 a 25.2 abc

MF-40 5 505 de -71 13.5 ab 27.5 ab

ET-35 4 856 de -74 14.8 a 27.6 ab

MF-25 4 793 de -74 14.1 a 28.3 a

MT-20 3 271 e -83 14.0 a 24.9 abc

AC6 - - 12.6 ab 26.5 abc1 Numbers followed by the same letter in a column are not different according to the Tukey test (P ≤ 0.05) in SAS.2 Number of vermiform R. similis nematodes/100 g root recovered from root system at time of evaluation (2 months after inoculation with 500 R. similis).3 Root weight: fresh weight of root system of plant at time of evaluation.4 Shoot weight includes the fresh weight of corm, pseudostem and leaves of plants at time of evaluation.5 Control inoculated with R. similis, but without fungal endophyte spores.6 AC: absolute control (no endophytes, no nematodes).

The predominant genera identified were Fusarium and Trichoderma (Table 2). Though more Chrysosporium, Candida and Trichotechium were recovered from roots from El Real farm, these genera are known saprophytes, and were not considered endophytic. Only Fusarium and Trichoderma isolates that demonstrated good growth in vitro were selected for screening.

The screening of 12 selected endophytes revealed that all had some antagonistic effects on the reproduction of R. similis in the roots of ‘Grande naine’ plantlets over a two-month period (Table 3). Additionally, most endophytes positively influenced plant growth, as indicated by increased root and shoot fresh weight. The four most effective endophytes, the Fusarium isolates MF-25 and MF-40, and the Trichoderma isolates MT-20 and ET-35, were later identified as Fusarium oxysporum and Trichoderma atro-viride isolates, by Dr Helgard Nirenberg from the BBA, in Berlin, Germany.


them, Fusarium and Trichoderma spp. were most commonly encountered endophytes. All those endophytes screened for antagonistic effects were effective in reducing or suppressing R similis in banana roots. F. oxysporum and T. atroviride isolates exhibit strong antagonistic activity towards R. similis. This is the first time this type of beha-viour has been observed for Trichoderma, confirming its biological control potential in banana. We were able to establish suppressiveness in micropropagated banana plantlets by inoculating them with conidia of these fungi before planting, indicating that this enhancement of banana planting material can be an effective means of attaining biolo-gical system management in the field. While micropropagated plants provide producers with guaranteed nematode-free planting material, they are also free of non-pathogenic microorganisms such as beneficial endophytic microorganisms (Pereira et al. 1999). This makes them an easy target for soil-borne diseases and nematodes. Therefore, the use of endophytic fungi as biocontrol agents of R. similis and other plant-parasitic nematodes has great potential on such planting material. By inoculating micropropa-gated plants with endophytes before planting, they are provided with tailored and “on site” protection. The use of endophytes also eliminates a major problem associated with the use of soil microbial biocontrol agents to control nematodes: the inoculation of “problem” soils with biocontrol organisms (Rodríguez-Kábana 1991). By placing the biocontrol agent directly inside the root, it is at least partially protected from other microorganisms present in the soil. This ensures a high density of biocontrol acti-vity. Nevertheless, the non-pathogenic nature of the promising biocontrol endophytes toward banana and other crop species has to be investigated before field experiments can be undertaken.

ReferencesCurl E.A. & B. Truelove. 1986. The Rhizosphere. Advanced Series in Agricultural Sciences, Springer Verlag, Berlin.

Dicklow M.B., N. Acosta & B.M. Zuckermann. 1993. A novel Streptomyces species for controlling plant-parasitic nematodes. Journal of Chemical Ecology 19:159-173.

Kerry B.R., D. H. Crump & L.A. Mullen. 1982. Studies of the cereal cyst nematode, Heterodera avenae, under conti-nuous cereals, 1974-1978. I. Plant growth and nematode multiplication. Annals of Applied Biology 100:477-87.

Niere B.I., P.R. Speijer, C.S. Gold & R.A Sikora. 1999. Fungal endophytes from bananas for the biocontrol of Radopholus similis. Pp. 313-318 in Mobilizing IPM for sustainable banana production in Africa (E.A. Frison, C.S. Gold, E.B. Karamura & R.A. Sikora, eds). INIBAP, Montpellier, France.

Pereira J.O., M.L. Carneiro Vieira & J.L. Azevedo. 1999. Endophytic fungi from Musa acuminata and their reintroduc-tion into axenic plants. World Journal of Microbiology & Biotechnology 15:37-40.

Pocasangre L., R.A. Sikora, V. Vilich & R.P. Schuster. 2000. Survey of banana endophytic fungi from Central America and screening for biological control of Radopholus similis. Acta Horticulturae 531:283-289.

Rodríguez-Kábana R. 1991. Biological control of plant parasitic nematodes. Nematropica 21:111-122.

Sikora R.A. 1992. Management of the antagonistic potential in agricultural ecosystems for the biological control of plant parasitic nematodes. Annual Review of Phytopathology 30:245-270.

Sikora R.A. 1997. Biological system management in the rhizosphere: An inside-out/outside-in perspective. Journal of Plant Protection 62:105-112.

Sikora R.A. 2002. Strategies for biological system management of nematodes in horticultural crops: fumigate, confuse or ignore them. Journal of Plant Protection 67:5-18.

Sikora R.A. & S. Reimann. 2004. Suppressive soils, the edge of chaos and multi-trophic strategies for biocontrol of pests and diseases in soil ecosystems. Multitrophic Interactions in Soil IOBC wprs Bulletin 27:251-258.


Sikora R.A. & R.P. Schuster. 1998. Novel approaches to nematode IPM. Pp. 127-136. in Proceeding of a workshop on Mobilizing IPM for sustainable banana production in Africa. (E.A. Frison, C.S. Gold, E.B. Karamura & R.A. Sikora, eds). INIBAP, Montpellier, France.

Speijer P.R. & D. De Waele. 1997. INIBAP Technical Guidelines. 1. Screening of Musa germplasm for resistance and tolerance to nematodes. INIBAP, Montpellier, France.

Speijer P.R. & C.S. Gold. 1996. Musa root health assessment: a technique for the evaluation of Musa germplasm for nematode resistance. Pp. 62-78 in New Frontiers in Resistance Breeding for Nematodes, Fusarium and Sigatoka. (E.A Frison, J.P. Horry & D. De Waele, eds). INIBAP, Montpellier, France.

Vilich V. & R.A. Sikora. 1998. Diversity in soil-borne microbial communities: a tool for biological system management of root health. Pp. 1-15 in Plant-Microbe Interactions and Biological Control. (G.J. Boland & L.D. Kuykendall, eds). Marcel Dekker Inc., New York, USA.

The problem of banana root deterioration and its impact on production178 Effect of arbuscular mycorrhizal fungi and other rhizophere microorganisms... C.A. Gauggel et al. 179M.C. Jaizme-Vega et al.

Effect of arbuscular mycorrhizal fungi (amf) and other rhizosphere microorganisms on development of the banana root systemM.C. Jaizme-Vega1, A.S. Rodríguez-Romero1 and M.S. Piñero Guerra1

AbstractDespite its influence on plant growth, root architecture has not been studied as much as it should be. Root architecture and morphology provide a lot of useful information about plant species and their ability to take up nutrients from the soil. Energy flow can also be examined in these studies. Several soil factors, such as nutrients and microorganisms, greatly affect root development. Soil microbiota interactions, especially with those organisms that colonize the rhizosphere, have also been reported to affect plant health and soil quality. Microbiota help the host plant under limiting conditions provoked by abiotic and/or biotic factors. Arbuscular mycorrhizal fungi (AMF) and plant-growth promoting bacteria (PGPBs) are among these beneficial rhizosphere microbes. Mycorrhizal symbiosis significantly improves plant nutrition under low fertility soil conditions. Mycorrhizal hyphae are more efficient than roots alone in nutrient uptake, since they increase the root absorption area. Recently, it has been reported that AMF are even able to change root architecture. Bananas (Musa acuminata Colla AAA) show a great ability to establish symbiotic associations with both AMF and PGPBs. Two experiments were conducted: one to evaluate the effect of AMF (Glomus intraradices) on banana root architecture of micropropagated ‘Grande Naine’ plantlets, in an 80-day sequential study. The second experiment was conducted to determine the effect of the interaction between AMF and PGPBs on root morphology. In this case, three micropropagated banana cultivars (‘Grande Naine’, ‘Dwarf Cavendish’ and ITC #1283 from the INIBAP gene bank), were inoculated with Glomus manihotis and with a mix of rhizosphere Bacillus spp. From these experiments it can be concluded that the rhizosphere microorganisms studied (AMF and PGPBs), promote positive banana root system development that leads to an improvement in plant nutrition and health. Thus, these microorganisms could be used in future alternative biotechnologies for banana production systems.

Resumen - Efecto de los hongos micorrícicos arbusculares (HMA) y de otros microorganismos de la rizosfera en el desarrollo del sistema radical del bananoLa arquitectura de las raíces ha sido poco estudiada en los trabajos de fisiología vegetal, a pesar de su gran importancia para el desarrollo de las plantas. Conocer la arquitectura y morfología de un sistema radical, nos proporciona información útil acerca de la especie en estudio, y su capacidad para absorber nutrientes del suelo. El flujo de energía también puede ser espinado en estos estudios. Existen muchos factores del suelo, tales como los nutrientes y los microorganismos que influyen en el desarrollo de la raíz. Las interacciones microbianas en la rizosfera inciden notablemente en la salud vegetal y calidad del suelo. Las interacciones de la microbiota del suelo, especialmente con los organismos que colonizan la rizosfera, han sido reportadas influenciando la salud y calidad de los suelos. La microbiota ayuda a la planta hospedera a adaptarse a situaciones de estrés de tipo hídrico o nutricional e incluso, a aquéllas originadas por patógenos del sistema radical. Los hongos formadores de micorrizas arbusculares (MA) y las bacterias promotoras del crecimiento vegetal (PGPBs), pertenecen al grupo de microorganismos que cohabitan en el entorno rizosférico. La simbiosis micorrícica mejora significativamente la nutrición en suelos de baja fertilidad. Las hifas de estos hongos son más eficaces en la captación de nutrientes que las propias raíces, incrementando el área de absorción radical. Recientemente se sabe que los hongos micorrícicos son capaces de alterar

1 Dpto. Protección Vegetal, Instituto Canario de Investigaciones Agrarias Apdo. 60. 38200 La Laguna, Tenerife, España. e-mail: [email protected]


la arquitectura de la raíz y propiciar una mayor eficiencia para la toma de nutrientes en las plantas micorrizadas. El banano (Musa acuminata Colla AAA) muestra una gran capacidad para beneficiarse de la simbiosis micorrícica y de la actividad de ciertas PGPBs. Se desarrollaron dos experimentos: uno para comprobar el efecto de los hongos MA Glomus intraradices ,sobre la arquitectura radical del banano (vitroplantas de ‘Gran Enana’), evaluando el efecto a lo largo de 80 días. El segundo experimento, tuvo como objetivo determinar el efecto de la interacción entre hongos MA y PGPBs sobre la morfología radicalen tres cultivares de banano micropropagado (‘Gran Enana’, ‘Pequeña Enana’ y el ITC #1283 del INIBAP), tratados con inóculo de Glomus manihotis y un cóctel de la bacteria rizosférica Bacillus spp. Ambos trabajos nos permiten concluir que los microorganismos rizosféricos estudiados (hongos MA y PGPBs), promueven el desarrollo del sistema radical del banano, lo cual mejora la nutrición y salud de las plantas. Se propone su utilización en los sistemas de producción del banano como una alternativa biotecnológica de futuro.

IntroductionDespite its influence on plant growth, root architecture has not been studied as much as it should be. Root architecture and morphology provide a lot of useful information about plant species and their ability to take up nutrients from the soil. Energy costs, involved in this nutrient uptake as a consequence of environmental adaptation, can also be examined in these studies. Indeed, the successful establishment of some crops depends on the adequate development of the root system of the plant. Survival rates during the post vitro phase of micropropagated material can often be reduced due to poor root systems (Hooker et al. 1994).

Among the environmental factors involved in root development (for example, soil structure, temperature, water, nutrient availability), communities of soil microbiota must be considered. Soil microbiota interactions, especially with those organisms that colonize the rhizosphere (the soil zone influenced by roots through the release of substrates that affect microbial activity), have also been reported to affect plant health and soil quality. Root-associated microbiota help the host plant under limiting condi-tions caused by abiotic (water, nutrition) and/or biotic (soil-borne pathogens) factors. Arbuscular mycorrhizal fungi (AMF) and plant-growth promoting bacteria (PGPBs) are among these beneficial rhizosphere microbes. Mycorrhizal symbiosis significantly improves plant nutrition under low soil fertility. Mycorrhizal hyphae are more efficient than roots alone in nutrient uptake, especially of elements with low mobility in soil such as phosphorus (P). Some studies have also reported changes in phytohormone balance (Drüge and Schönbeck 1992). Recently, it has been reported that AMF are even able to change root architecture. These changes lead to more efficient nutrient uptake in mycorrhizal plants (Hooker and Atkinson 1992).

Additionally, some rhizosphere bacteria promote plant development and seed germina-tion. This ability should be given serious attention, because micropropagated material is being used increasingly in commercial crops. PGPBs, as well as AMF, are able to enhance biomass production during the first stages of crops and can contribute to plant adaptation from the in vitro to the post vitro phase (Carletti 2000). Direct mechanisms such as nutrient solubility or fixation and siderophore or phytohormone production have been reported for some bacterial strains. Siderophores are considered relevant factors in plant nutrition, since they can act as an Fe3+-chelating agent. It is known that some auxin-type phytohormones (indol-acetic acid, IAA) produced by PGPB strains,


promote lateral and primary root elongation and production, with positive effects on root system development (Kloepper et al. 1989 & 1991).

Since they share common microhabitats, AMF and PGPBs must interact during the colonization process and/or as rhizosphere microorganisms (Barea 1997). It is accep-ted that soil microbial communities, in general and especially PGPBs can have a great influence on the establishment of mycorrhizal symbiosis. In the same way, AMF can affect PGPB populations in roots. Thus, microbial interactions should be taken into account in studies on both the single and cooperative effects of these organisms on root systems (Barea 1997, 2000). Banana (Musa acuminata Colla AAA), is a monocotyle-donous herbaceous species, that shows a great ability to establish mycorrhizal symbio-sis (Jaizme-Vega et al. 1991, 1998, 2002, Rizzardi 1990, Declerck et al. 1994, Yano-Melo et al. 1999). Banana-PGPB associations have also been reported as beneficial (Rodríguez-Romero et al. 2003, Jaizme-Vega et al. 2004). However, despite the great number of studies on banana, there are not many references concerning the effect of rhizosphere microbiota on banana root architecture (Jaizme-Vega et al. 1994; García-Pérez and Jaizme-Vega 1997) and their consequences on plant growth and health.

This study examines the effects of AMF and PGPBs and their interaction on the banana root system. Two different experiments were conducted. The first determines the effect of AMF on root morphology of ‘Grande Naine’. The second analysed the interaction of both microorganisms (AMF and PGPBs), on three banana cultivars (‘Grande naine’, ‘Dwarf Cavendish’ and a somaclonal variant of SH 3436-9 with ITC1283).

Materials and methodsExperiment 1: AMF – ‘Grande naine’Host plants, AM fungi and experimental conditionsMicropropagated banana (Musa acuminata Colla) cv. ‘Grande naine’ plantlets were used. After washing with fresh water to remove agar, they were transplanted from vitro flasks to two 24 L black polyethylene plastic containers (20 plants/container) filled with a sterilized substrate (3:1) of quartz sand and soil with a low concentration of phos-phorus (12.7 mg/kg Olsen). Half the plants were inoculated with Glomus intraradices Schenck & Smith, cultured under Sudan grass (Sorghum bicolor L. var. sudanense Moench). AMF inoculum consisted of 25 g per container of rhizosphere soil containing external mycelium and infected root fragments obtained from pot cultures. Plants were maintained under greenhouse conditions (24ºC, 70% RH), receiving 600 ml of nutrient solution without phosphorus, per week (Hewitt 1952).

Assessment of variables and analyses Four plants from each of the two treatments were sampled at 30, 40, 50, 65 and 80 days after transplanting and assessed for changes in biomass and morphological deve-lopment. At each harvest, mycorrhizal root colonization was evaluated microscopically after staining with 0.05% trypan blue in lactic acid (Phillips and Hayman 1970) using the procedure described by Koske and Gemma (1989). The percentage of root system colonization was quantified using the grid intersect method (Giovannetti and Mosse 1980). At each harvest the following parameters were determined: number and length


of adventitious roots, number and length of lateral roots, lateral root frequency (repre-sented by the number of roots of order n, divided by the length of roots of order n-1) and the percentage of roots present in each order. All data were compared by ANOVA and means were separated by Tukey’s test (P≤0.05). All statistical analyses were per-formed with Systat 7.0.

Experiment 2: AMF - PGPBs - ‘Grande naine’ / ‘Dwarf Cavendish’ / ITC1283

Host plants, AM fungi and plant growth promoting bacteriaMicropropagated banana plantlets from three cultivars (‘Grande naine’, ‘Dwarf Cavendish’ and a somaclonal variant of SH 3436-9 from Cuba with ITC1283) were used. After washing with fresh water to remove agar, plants were transplanted to 24 L black polyethylene (PE) plastic containers (24 plants/container) filled with a sterili-zed substrate (1:1:1) made of soil, volcanic ash and peat TKS1-Instant (Sphagnum-Torf Klasman-Deilmann GmbH, Germany). The AMF was Glomus manihotis, from Colombia, cultured under Sudan grass (Sorghum bicolor L. var. sudanense Moench), with a 78% colonization index. AMF inoculum consisted of 1.5 kg per container of rhizosphere soil containing external mycelium and infected root fragments obtained from pot-cultures. For each banana cultivar half plants were inoculated with AMF. A mix of Bacillus spp., three strains (INR 7, T4 and IN 937 b) isolated and identified by Dr. Kloepper (Alabama, USA), were kept in TSB (Tryptone Soy Broth) with 20% glycerol. Bacterial inoculums consisted of a sterilized saline (0.85% NaCl) suspension containing an equal amount of each strain. Colonies were obtained after culturing onto Petri dishes with TSA (Tryptone Soy Agar) for two weeks. Bacterial suspension, con-taining approximately 108 colony forming units (CFU)/ml was added 20 days after transplanting. For each cultivar and treatment (mycorrhizal and non mycorrhizal), 20 ml were added to half of the plants

Experimental conditions and assessment of variables Plants were kept under controlled conditions during the 60-day hardening phase (90% RH and 25ºC). During this period, plants were watered properly. At the end of the har-dening phase, plants were transferred to individual 3 L black PE pots filled with a steri-lized substrate (1:1:1) made of soil, volcanic ash and peat TKS1-Instant (Sphagnum-Torf Klasman-Deilmann GmbH, Germany). Plants were arranged in a random design under greenhouse conditions (75% RH and 30ºC). Fertilization was conducted following a programme similar to those of commercial nurseries for banana. Twice a month a foliar fertilizer, Wuxal Super AA-8-8-6 (3%), was added. Four assessments were made at 30, 85, 135 and 185 days after bacterial inoculation. At each assessment the following parameters were measured on 6 plants in each cultivar and treatment: number of adventitious roots and shoot fresh weight (g). Leaf area (cm2) was measu-red using a leaf area meter (LiCOR, Inc Lincoln, Nebraska, USA. Mod. Li-3100). The mycorrhizal colonization index was determined at each assessment. All data were compared by ANOVA and means were separated by Tukey’s test (P≤0.05). All statistical analyses were performed with Systat 7.0.


ResultsExperiment 1: AMF - ‘Grande naine’All the micropropagated banana plants used in this experiment were similar in size at inoculation. Fresh weight of the aerial part of the mycorrhizal plants increased in most of the harvests. Differences between the inoculated and control plants were more marked at 30 and 80 days, when the values were significantly higher in G. intraradices inoculated plants (Figure 1). Mycorrhizal root infection was absent from control plants, but well established in the mycorrhizal plants. At the first assessment (30 days after inoculation) mycorrhizal colonization was 70%, at the second one (20 days later) colo-nization was 80%, and increased to 100% from 50 days after inoculation (Figure 2).

The root system of micropropagated Musa plants consists of a series of adventitious roots that develop directly from the corm (subterranean structure). These roots branched in first order laterals, from which second order laterals developed. Mycorrhizal banana plants showed a denser root system than the control plants. Adventitious root number increased over time in both mycorrhizal and non-treated plants, although values of mycorrhizal plants were always significantly higher (Figure 3). In the case of adven-titious root length, mycorrhizal plants showed lower values than non-treated plants, between the second and the third assessment, although differences were not significant (Figure 4). The degree of branching of adventitious roots (the number of branches per primary axis) increased over time in AMF treatments. At 30, 65 and 80 days post inoculation there were significant differences between treatments. Adventitious root branching was highest for the AMF treatment at day 30. The intensity of branching of first order laterals was not determined in this study (Figure 5).

a

baaab

aaa

a

a

0

2

4

6

8

10

12

0 20 40 60 80 100

g

Control LINR

0

20

40

60

80

100

120%

0 20 40 60 80 100

Control LINR

Figure 2: Variation with time of root infection percentage in Glomus intraradices (LINR) in micropropagated banana plants cv. ‘Grande naine’.

Figure 1: Effect of Glomus intraradices (LINR) on shoot fresh weight of micropropagated banana plants cv. ‘Grande naine’. For each time, the data followed by the same letter are not significantly different according to Tukey’s test (P≤0.050).


Experiment 2: AMF - PGPBs- ‘Grande naine’, ‘Dwarf Cavendish’, ITC1283Combined inoculation (G. manihotis+ Bacillus spp.) of plants resulted in an increase in plant development compared with single inoculation of either AMF or PGPB. All physical parameters were higher in plants treated with both microorganisms. However, for some physical parameters single inoculation with Bacillus spp. or AMF were statistically identical with non-treated plants. Nevertheless, each banana cultivar showed a different trend.

In the case of ‘Grande naine’ cultivar, significant differences were highest 135 days after bacteria inoculation at which time, plants treated with AMF and PGPB,

showed significant increases in physical development (Figure 6). For ‘Dwarf Cavendish’ plants, highest significant differences in adventitious root number were detected 135 days after Bacillus application when Bacillus-treated plants showed a greater number of adventitious roots. Significant increases in leaf area and fresh weight of the aerial parts were also registered for this treatment at the first and second assessments (Figure 7). INIBAP selection ITC1283 showed a generally good response after inoculation with both organisms. In this case, significant increases in plant development (leaf area and shoot fresh weight) could be detected at the first assessment. Plants from this cultivar colonised with G. manihotis were able to increase adventitious root number, leaf area and shoot fresh weight at the end of the experiment (Figure 8).

0 20 40 60 80 100

Control LINR

Days after AMF inoculation

b

aba

b

a

a aa

a

0

1

2

3

Figure 5: Variation with time of adventitious root bran-ching in Glomus intraradices (LINR) and control micropro-pagated banana plants cv. ‘Grande naine’. For each time, the data followed by the same letter are not significantly different according to Tukey’s test (P≤0.050).

a a aa

a

a

a a

a

a

0

2

4

6

8cm

0 20 40 60 80 100

Control LINR


Figure 4: Variation with time of adventitious root length in Glomus intraradices (LINR) and control micropropa-gated banana plants cv. ‘Grande naine’. For each time, the data followed by the same letter are not significantly different according to Tukey’s test (P≤0.050).

b b

b b

b

a

aaa

a

0

2

4

6

8

10

12

0 20 40 60 80 100

Control LINR


Figure 3: Variation with time of adventitious root number in Glomus intraradices (LINR) and control micropropa-gated banana plants cv. ‘Grande naine’. For each time, the data followed by the same letter are not significantly different according to Tukey’s test (P≤0.050).


P=0.656P=0.796

P=0.050

P=0.379

0

10

20

30

40

30 85 135 185

(a)

P=0.009P=0.993P=0.301

P=0.383

0

1000

2000

3000

4000

30 85 135 185

cm2(b)

P=0.910P=0.799 P=0.001

P=0.332

0

100

200

300

400

30 85 135 185

Days after bacterial inoculation



g

Control PGPB LMNH LMNH+PGPB

(c)

Figure 6: Effect of combined inoculation of Glomus manihotis (LMNH) and Bacillus (PGPB) on (a) adventitious root num-ber, (b) leaf area and (c) shoot fresh weight on ‘Grande naine’ micropropagated plants.


0

10

20

P=0.755

P=0.444

P=0.032

P=0.864

P=0.038P=0.005

P=0.576

P=0.492

P=0.088 P=0.006

P=0.666

P=0.016

30

30 85 135 185

(a)

0

1000

2000

3000

30 85 135 185

cm2(b)

0

100

200

300

400

30 85 135 185




g


(c)

Figure 7: Effect of combined inoculation of Glomus manihotis (LMNH) and Bacillus (PGPB) on (a) adventitious root num-ber, (b) leaf area and (c) shoot fresh weight on ‘Dwarf Cavendish’ micropropagated plants.


0

10

20

P=0.045

P=0.019

P=0.191

P=0.541

P=0.473

P=0.888

P=0.846

P=0.700

P=0.937

P=0.003

P=0.006

P=0.001

30

40

30 85 135 185

(a)

0

1000

2000

3000

30 85 135 185

cm2(b)

0

100

200

300

400

30 85 135 185




g


(c)

Figure 8: Effect of combined inoculation of Glomus manihotis (LMNH) and Bacillus (PGPB) on (a) adventitious root num-ber, (b) leaf area and (c) shoot fresh weight on ITC1283 micropropagated plants.


It can be said in general, that all cultivars, except ITC1283, increased adventitious root number after the inoculation with both microorganisms. Concerning mycorrhizal colonization, all banana cultivars showed an acceptable root colonization index. No significant difference due to bacterial inoculation could be detected (Figures 9, 10 and 11).

DiscussionMycorrhizal banana plants had root systems with more numerous and branched adventitious roots than those of the control plants. No significant differences between adventitious root length of mycorrhizal and control plants were observed at the first assessment, 30 days after transplanting, when the degree of colonization was still relati-vely low (about 30%). Later on, when infection increased, adventitious root length was always lower in mycorrhizae, and significantly lower at 40 and 50 days after transplan-ting. At 65 days after transplanting, mean adventitious root length was about the same in mycorrhizal and control plants. The lack of difference in adventitious root length between mycorrhizae and control plants at the last sample, 80 days after transplanting, could be partly explained by senescence processes in both treatments.

The reduction in length of adventitious roots was accompanied by an increase in their number, as observed in another endomycorrhizal system in which a monocotyledon was involved i.e. Allium porrum + Glomus E3, Ornithogalum umbellatum + G. fasciculatum (Berta et al. 1990, 1993). Lateral root length did not change in mycorrhizal banana plants, even if there was a trend towards an increase. The same occurred in Platanus acerifolia (Tisserant et al. 1992), in Populus sp. and Prunus cerasifera. Lateral roots were longer in mycorrhizal than in the control plants (Hooker et al. 1992, Berta et al. 1995) whereas in Vitis vinifera and AIIium porrum lateral roots were shorter in the mycorrhizae plants (Schellenbaum et al. 1991, Berta et al. 1993).

The most important effect of AMF infection on the root system of banana plants was an increase of adventitious root branching. This increase, together with the increase in number of adventitious roots, led to a denser root system, as observed in other endo-mycorrhizal systems, though effects may be different (Berta et al. 1993). It is generally assumed that a denser root system has a greater absorbing power than an elongated one and is typical of plants growing in nutrient rich soils (Glinski and Lipiec 1990). The extensive network of external mycelium, with its absorbing power and explorative functions, in addition to the denser root system, could improve the “growth effect” cha-racteristic of mycorrhizal plants. Moreover, a very branched root system is particularly useful for banana plants, as they are easily uprooted by winds, which are a frequent problem in many cropping areas of the world.

Inoculation with both microorganisms lead to a positive effect on banana growth in all cases. Significant increases were detected compared with single inoculated treatments or non-treated plants. It can be assumed then, that under our experimental conditions, this association should be considered a case of synergism. Each banana cultivar showed a particular trend (Table 1). ‘Grande naine’ plantlets showed significant effects due to the combined treatment, at 135 days after inoculation. Those effects were mainly detec-ted in adventitious root number. ‘Dwarf Cavendish’ plants had a similar response to the ‘Grande naine’ cultivar. On the other hand, ITC1283 showed significant increases in


30

80

60

40

20

0

80

60

40

20

0

80

60

40

20

0

85 135 185

(a)

30 85 135 185

(b)

30 85 135 185




LMNH LMNH+PGPB

(c)

%

%

%

Figure 11: Variation with time of per-centage of root infection in Glomus manihotis (LMNH) inoculated and con-trol banana plants selection of ITC1283.

Figure 10: Variation with time of percentage of root infection in Glomus manihotis (LMNH) inoculated and control banana plants cv. ‘Dwarf Cavendish’.

Figure 9: Variation with time of percentage of root infection in Glomus manihotis (LMNH) inoculated and control banana plants cv. ‘Grande naine’.


adventitious root number at the end of the experiment and just for plants with mycor-rhizae alone.

Our results did not detect a direct effect of PGPBs on banana root morphology. Significant increases in adventitious root number only occur in mycorrhizal treated plants (mycorrhizae alone or combined with PGPB) in all cultivars. There are few references concerning the effects of PGPB on root morphology. Williamson and Jones (1973) found certain changes in root morphology of maize after inoculation with Bacillus subtilis. Other works (Loper and Schroth 1986) have detected a positive effect on root hair elongation in sugar beet, after inoculation with rhizosphere bacteria. Recently, Lippmann et al. (1995) have described the beneficial effects of some IAA-producing Pseudomonas strains on root elongation and lateral root production in maize seedlings. These effects were consequently related to plant shoot growth.

Microbial relationships in the rhizosphere have high specificity (Kloepper 1996). Interactions between beneficial microorganisms also confirm this specificity (Azcón 1989, Paulitz and Linderman 1989). Because most previous work concerns other crops, comparisons cannot be made with the same plant species. Other authors have already mentioned the beneficial use of these associations (Dihillion 1992, Singh and Kapoor 1998, Ravnskov and Jakobsen 1999, Ratti et al. 2001). Most described increases in plant biomass and nutrient contents (Dihillion 1992, Singh and Kappor 1998, Potty and Pillai 1996). The potential use of both AMF and P-solubilizing bacteria to reduce phosphorus fertilizer inputs has also been considered (Potty and Pillai 1996). In other cases, the diversity of the effects of different microbial associations has been mentioned (Toro et al. 1996, Ravnskov and Jakobsen 1999).

Table 1. The significant (p<0.05) responses of young banana plants, grown post vitro, to either the inoculation of Glomus manihotis (LMNH), or the addition of a mix of Bacillus spp. plant growth promoting bacteria (PGPB), or to a combination of G. manihotis and Bacillus spp. (LMNH+PGPB) during the first 185 days after inoculation. Cultivar Root number/plant Leaf area Shoot fresh weight

30 days after bacterial inoculation

Grande naine - - -Dwarf Cavendish - LMNH+PGPB -INIBAP ITC1283 - LMNH+PGPB LMNH+PGPB


Grande naine - - -Dwarf Cavendish - PGPB PGPBITC1283 - - -


Grande naine LMNH+PGPB LMNH+PGPB LMNH+PGPBDwarf Cavendish LMNH+PGPB - -ITC1283 - - -


Grande naine - - -Dwarf Cavendish - - LMNHITC1283 LMNH LMNH LMNH- = no significant response in this combination of age and cultivar.


There are also some references concerning the effect of PGPBs on AMF (Burla et al. 1996, Germida and Walley 1997, Attia 1999). Some describe negative effects on mycorrhizal colonization development (Burla et al. 1996, Germida and Walley 1997). Our results confirm what Attia (1999) described: the absence of any negative effect on AMF, since the colonization root indices of double inoculated plants were identical to those with just AMF. Results from this experiment show that combined inoculation has positive effects on plant development and adventitious root number. However, due to the lack of references concerning the effect of PGPB on the banana root system, further experiments should be conducted to achieve a better approach.

ReferencesAttia M. 1999. The efficiency improvements of mineral fertilizers used and maize yield by arbuscular mycorrhizal

fungus and plant growth-promoting rhizobacteria. Annals of Agricultural Science, Cairo 44:41-53.

Azcón R. 1989. Selective interactions between free-living rhizospheric bacteria and vesicular-arbuscular mycorrhizal fungi. Soil Biology and Biochemistry 21:639-644.

Barea J.M. 1997. Mycorrhiza/bacteria interactions on plant growth promotion. Pp.150-158 in Plant growth-promoting rhizobacteria, present status and future prospects. (A Ogoshi, L Kobayashi, Y Homma, F Kodama, N Kondon and S Akino, eds). OECD, Paris.

Barea J.M. 2000. Rhizosphere and mycorrhiza of field crops. Pp. 110-125 in Biological resource management: con-necting science and policy (OECD). (JP Toutant, E Balazs, E Galante, JM Lynch, JS Schepers, D Werner and PA Werry, eds). INRA Editions and Springer.

Berta G., A. Fusconi & T. Trotta. 1993. VA mycorrhizal infection and the morphology and function of root systems. Environmental and Experimental Botany 33:159-173.

Berta G., A. Fusconi, T. Trotta & S. Scannerini. 1990. Morphogenetic modifications induced by the mycorrhizal fungus Glomus strain E3 in the root system of Allíum porrum L. New Phytologist 114:207-215.

Berta G., A. Trotta, A. Fusconi , S.E. Hooker, M. Munro, O. Atkinson, M. Giovannetti, S. Moriní, P. Fortuna, B. Tisserant, V. Gianinazzi-Pearson & S. Gianinazzi. 1995. Arbuscular mycorrhizal induced changes to plant growth and root system morphology in Prunus cerasifera. Tree Physiology 15:281-293.

Burla M., M. Govede, F.J. Schwinn & A. Wiemken. 1996. Influence of biocontrol organisms on root pathogenic fungi and on the plant symbiotic microorganisms Rhizobium phaseoli and Glomus mosseae. Zeitschrift für Pflanzenkrankenheiten und Pflanzenschutz 103:156-163.

Carletti S. 2000. Use of plant growth-promoting rhizobacteria in plant micropropagation. Auburn University Web Site: http://www.ag.auburn.edu/argentina/pdfmanuscripts/carletti.pdf.

Declerck S., B. Devos, B. Delvaux & C. Plenchette. 1994. Growth response of micropropagated banana plants to VAM inoculations. Fruits 49:103-109.

Dihillion S.S. 1992. Dual inoculation of pre-transplant stage Oryza sativa L. plants with indigenous vesicular-arbuscular mycorrhizal fungi and fluorescent Pseudomonas spp. Biology and Fertility of Soils 13:147-151.

Drüge U. & F. Schönbeck. 1992. Effect of vesicular-arbuscular mycorrhizal infection on transpiration, photosynthesis and growth of flax (Linum usitatissimum L.) in relation to cytokinin levels. Journal of Plant Physiology 141:40-48.

García Pérez, J. & M.C. Jaizme-Vega. 1997. Influence of infection by mycorrhizal fungus Glomus intraradices on plant growth and root development of Grande Naine banana. Abstract. p.155. International Symposium on Banana in the Subtropics. 10-14 Nov. Puerto de La Cruz, Tenerife, Canary Islands, Spain.

Germida J.J. & F.L. Walley. 1997. Plant growth promoting rhizobacteria alter rooting patterns and arbuscular mycor-rhizal fungi colonization of field-grown spring wheat. Biology and Fertility of Soils 23:113-120.

Giovanetti M. & B. Mosse. 1980. An evaluation of techniques for measuring vesicular-arbuscular mycorrhizal infection in roots. New Phytologist 84:489-500.

Glinski J. & J. Lipiec. 1990. Soil physical conditions and plant roots. CRC Press, Boca Raton, FL

Hewitt E.S. 1952. Sand and water culture methods used in the study of plant nutrition. (Technical communication 22). Farnham Royal Commonwealth Agricultural Bureau, Farnham, UK.


Hooker J.E. & U. Atkinson. 1992. Application of computer-aided image analysis to studies of arbuscular endomycor-rhizal fungi effects on plant root system morphology and dynamics. Agronomie, Paris 12:821-824.

Hooker J.E., M.C. Jaizme-Vega & D. Atkinson. 1994. Biocontrol of plant pathogens using arbuscular mycorrhizal fungi. Pp. 191-200 in Sustainable agriculture and natural ecosystems. (S. Gianinazzi & H. Schüepp, eds). Birkhäuser Verlag, Basel, Switzerland.

Hooker J.E., M. Munro & U. Atkinson. 1992. Vesicular-arbuscular mycorrhizal fungi induced alteration in poplar root system morphology. Plant and Soil 145:207-214.

Jaizme-Vega M.C., G. Berta & S. Gianinazzi. 1994. Effect of Glomus intraradices on root system morphology of micropropagated banana plants. Abstract. P. 152 in Fourth European Symposium on Mycorrhizas, 11-14 July 1994. Granada. Spain.

Jaizme-Vega M.C., M. Esquivel Delamo, P. Tenoury Domínguez y A.S. Rodríguez-Romero. 2002. Efectos de la micorrización sobre el desarrollo de dos cultivares de platanera micropropagada. InfoMusa 11(1):25-28.

Jaizme-Vega M.C., V. Galán & J. Cabrera. 1991. Preliminary results of VAM effects of banana under field conditions. Fruits 46:19-22.

Jaizme-Vega M.C., B. Sosa Hernández & J.M. Hernández Hernández. 1998. Interaction of arbuscular mycorrhizal fungi and the soil pathogen Fusarium oxysporum f. sp. cubense on the first stages of ‘Grande Naine’ banana. Acta Horticulturae 490:285-295.

Jaizme-Vega M.C., Rodríguez-Romero, A.S. & Piñero Guerra, M.S. 2004. Potential use of rhizobacteria form the Bacillus genus to stimulate the plant growth of micropropagated banana. Fruits 59:83-90.

Kloepper J.W. 1996. Host specificity in microbe-microbe interactions. BioScience 46:406-409.

Kloepper J.W., R. Lifshitz & R.M. Zablotowicz. 1989. Free-living bacterial inocula for enhancing crop productivity. Trends in Biotechnology 7:39-44.

Kloepper J.W., R.M. Zablotowicz, B. Tipping & R. Lifshitz. 1991. Plant growth promotion mediated by bacterial rhizos-phere colonizers. Pp. 315-326 in The Rhizosphere and Plant Growth. (D.L. Keister & P.B. Cregan, eds). Kluwer Academic Publ. Dordrecht.

Koske R.E. & J.M. Gemma. 1989. A modified procedure for staining roots to detect VA mycorrhizas. Mycological Research 92:486-505.

Lippman B., V. Leinhos & H. Bergmann. 1995. Influence of auxin producing rhizobacteria on root morphology and nutrient accumulation of corps. I. Changes in root morphology and nutrient accumulation in maize (Zea Mays L.) caused by inoculation with indole-3-acetic acid (IAA) producing Pseudomonas and Acinetobacter strains or IAA applied exogenously. Angewandte Botanik 69:31-36.

Loper J.E. & M.N. Schroth. 1986. Influence of bacterial sources of indole-3-acetic on root elongation of sugar beet. Phytopathology 76:386-389.

Paulitz T.C. & R.G. Linderman. 1989. Interaction between fluorescent Pseudomonads and VA mycorrhizal fungi. New Phytologist 113:37-45.

Potty V.P. & S.W. Pillai. 1996. Interactions of Glomus microcarpum and phosphobacterium Bacillus sp. in the root zone of sweet potato and its role on growth and development. Pp. 247-251 in Tropical tuber crops: problems, prospects and future strategies. (G.T. Kurup, M.S. Palaniswami, V.P. Potty, G. Padmaja & S. Kabeerathumma, eds). Science Publishers, Inc. Lebanon, USA.

Phillips S.M. & U.S. Hayman. 1970. Improved procedures for clearing roots and staining parasitic and vesicular-arbus-cular mycorrhizal for rapid assessment of infection. Transaction of British Mycological Society 55:158-161.

Ratti N., S. Kumar, H.N. Verma & S.P. Gautam. 2001. Improvement in bioavailability of tricalcium phosphate to Cymbopogon martini var. motia by rhizobacteria, AMF and Azospirillum inoculation. Microbiological Research 156:145-149.

Ravnskov S. & I. Jakobsen. 1999. Effects of Pseudomonas fluorescens DF57 on growth and P uptake of two arbuscular mycorrhizal fungi in symbiosis with cucumber. Mycorrhiza 8:329-334.

Rizzardi V. 1990. Effect of inoculation with vesicular-arbuscular mycorrhizal fungi on the growth of micropropagated Musa acuminata clone ‘Grand Nain’. Rivista Agricoltura Subtropicale e Tropicale 84:473-484.

Rodríguez Romero A.S., M.S. Piñero-Guerra & M.C. Jaizme-Vega. 2003. Effect of the interaction between AMF Glomus manihotis and plant growth promoting bacteria (PGPB) from Bacillus genus on micropropagated banana plantlets. Abstract The Fourth International Conference on Mycorrhizae. ICOM4. 10-13 August 2003. Montreal. Canada.

The problem of banana root deterioration and its impact on production192 Effect of arbuscular mycorrhizal fungi and other rhizophere microorganisms... C.A. Gauggel et al. 193E. Fernandez et al.

Schellenbaum L., G. Berta, F. Ravolanirina, B. Tisserant, S. Gianinazzi & A.H. Fitter. 1991. Influence of endomy-corrhizal infection on root morphology in an micropropagated woody plant species (Vitís vinifera U.). Annals of Botany 68:135-141.

Singh S. & K.K. Kapoor. 1998. Effects of inoculation of phosphate-solubilizing microorganisms and an arbuscular mycorrhizal fungus on mungbean grown under natural soil conditions. Mycorrhiza 7:249-253.

Tisserant B., L. Schellenbaum, V. Gianinazzi-Pearson, S. Gianinazzi & G. Berta. 1992. Influence of infection by an endomycorrhizal fungus on root development and architecture in Platanus acerifolia. Alliona 30:171-181

Toro M., R. Azcón & R. Herrera. 1996. Effects on yield and nutrition of mycorrhizal and nodulated Pueraria phaseo-loides exerted by P solubilizing rhizobacteria. Biology and Fertility of Soils 21:23-29.

Williamson F.A. & R.G.W. Jones. 1973. The influence of soil microorganisms on growth of cereal seedlings and on potassium uptake. Soil Biology and Biochemistry 5:569-575.

Yano-Melo A.M., J.S. Orivaldo Jr, J.M. Lima-Filho, N.F. Melo & L.C. Maia. 1999. Effect of arbuscular mycorrhizal fungi on the acclimatization of micropropagated banana plantlets. Mycorrhiza 9:119-123.

The problem of banana root deterioration and its impact on production192 Effect of arbuscular mycorrhizal fungi and other rhizophere microorganisms... C.A. Gauggel et al. 193E. Fernandez et al.

Biological control of nematodes in bananaEmilio Fernández1, Jesús Mena2, Julián González3 and María Elena Márquez1

AbstractSeveral species of nematodes are associated with banana and plantain in Cuba; however, the most important are Radopholus similis, Meloidogyne incognita, Pratylenchus coffeae and Helicotylenchus multicinctus. These parasites have been dealt with for several years using integrated management systems. These systems include legal, cultural, genetic, biological and even chemical measures that have been adapted to the different development stages in this crop.

Biological control has shown to be an effective alternative that can be combined with others within an Integrated Management System. The base for biological control is the use of fungi such as Paecylomices lilacinus and bacteria such as Bacillus thuringiensis var. kurstaki. More recently, Corynebacterium paurometabolum bacterium has been added to this group as well as mycorrhiza from the Glomus genus.

P. lilacinus was the first bioregulator of banana parasitic nematodes researched in the country. Its effectiveness was demonstrated under semicontrolled, in vitro conditions, and in the field. B. thuringiensis var. kurstaki (LBT 3 strain) is a bacterium effective against insects that has also nematicidal effects, mainly against R. similis. It was possible to determine that its applications under pot and field conditions significantly reduced nematode attack. Evaluation of its effectiveness in more than 185 ha of production fields revealed average reductions of 87% in plantations with different populations. Studies with C. paurometabolum (C-294 strain) showed that it has effects on eggs and juveniles, manifested both under laboratory and field conditions. Reduction of populations under controlled conditions exceeded 85%, while in the field, R. similis populations started to decrease after the application and stayed low during the experiment, showing results comparable with the nematicide used as control.

With regards to mycorrhiza, early inoculation of in vitro plants at the hardening stage, with several species of the Glomus genus, minimized damage caused by R. similis and M. incognita.

Resumen - Control biológico de nematodos en bananoVarias especies de nematodos se encuentran asociadas al banano y el plátano en Cuba, no obstante las más importantes son Radopholus similis, Meloidogyne incognita, Pratylenchus coffeae y Helicotylenchus multicinctus. Estos parasitos se han combatido desde hace varios años, mediante sistemas de manejo integrado. Estos incluyen medidas legales, culturales, genéticas, biológicas e incluso químicas que se han adecuado a las distintas fases de desarrollo del cultivo.

El control biológico ha demostrado ser una alternativa efectiva, que permite ser combinada con el resto de las que componen el Sistema de Manejo Integrado. La base para el sistema de control biológico es el empleo de hongos como Paecylomices lilacinus y bacterias como Bacillus thuringiensis var. kurstaki. De forma mas reciente han entrado en este grupo la bacteria Corynebacterium paurometabolum (en fase acelerada de introducción) y micorrizas del género Glomus.

1 Instituto de Investigaciones de Sanidad Vegetal (INISAV), Ciudad Habana, Cuba. e-mail: [email protected]; [email protected]; 2 Centro de Ingeniería Genética y Biotecnología (CIGB). Camaguey. Cuba. e-mail: [email protected]; 3 Instituto de Investigaciones de Viandas Tropicales (INIVIT), Villa Clara, Cuba e-mail: [email protected]

The problem of banana root deterioration and its impact on production194 Biological control of nematodes in banana C.A. Gauggel et al. 195E. Fernandez et al.

P. lilacinus fue el primer bioregulador de nematodos parásitos del banano que se estudió en el país. Su efectividad fue demostrada en condiciones in vitro, semicontroladas y de campo. B.thuringiensis var. kurstaki (cepa LBT 3), bacteria efectiva contra insectos tiene efectos nematicidas, principalmente contra R. similis. Se pudo determinar que su aplicación tanto en condiciones de macetas como de campo reducía significativamente los niveles de ataque de los nematodos. La valoración de su efectividad en mas de 185 ha de campos de producción, reveló reducciones de 87% como promedio, en plantaciones con diferentes niveles poblacionales. Los estudios con C. paurometabolum (cepa C-294 ) revelaron que tiene efectos sobre huevos y juveniles, que se manifestaron tanto en condiciones de laboratorio como de campo. La reducción de los niveles poblacionales en condiciones controladas superó el 85%, mientras que a nivel de campo las poblaciones de R. similis comenzaron a disminuir a partir de la aplicación y se mantuvieron bajas durante todo el ensayo, con resultados comparables a la sustancia nematicida utilizada como control.

Con relación a las micorrizas, se ha determinado que la inoculación temprana de varias especies del género Glomus en la fase de endurecimiento de las vitroplantas, permite atenuar el daño por R. similis y M. incognita.

IntroductionSeveral species of nematodes are associated with banana and plantain in Cuba. The most important are Radopholus similis, Meloidogyne incognita, Pratylenchus coffeae and Helicotylenchus multicinctus, which have a varied distribution in Cuba.

These parasites have been controlled for several years using integrated management systems that combine preventive and curative alternatives in a harmonious manner. These systems include legal, cultural, genetic, biological and even chemical measures, adapted to the different developmental stages present in the crops. Some examples are: limited chemical nematicides treatments on commercial banana plantations, which were used until the end of the eighties when they gave way to other environmen-tally friendly practices; massive in vitro plant production as healthy planting material; and more recently, utilization of new black Sigatoka (Mycosphaerella fijiensis) and R. similis resistant cultivars developed by the Fundación Hondureña de Investigación Agrícola (FHIA) in Honduras.

Biological control has been shown to be an effective alternative that can be combined with other measures. The base for biological control has been the use of the fungus Paecilomyces lilacinus and the bacterium Bacillus thuringiensis var. kurstaki. More recently, the bacterium Corynebacterium paurometabolum has been added to this group as well as mycorrhizae from the Glomus genus. This paper deals with the main results obtained using biological control over several years in Cuba.

Materials and methodsPaecilomyces lilacinusThis fungus was the first biocontrol agent studied in the country. Its effectiveness was evaluated under in vitro and semi-controlled conditions and also in the field. In vitro spore suspensions were tested at different concentrations (105 to 109 conidia/g), using small Petri dishes with egg masses of M. incognita and larvae and adults of R. similis. In vitro plants, planted individually in plastic bags filled with sterilized soil, were used under semi-controlled conditions. Plants were first infected with 10 g of the bioproduct


of P. lilacinus and one week later, 1500 Meloidogyne juveniles were added per plant. Root evaluation was performed using Taylor and Sasser’s method (1978).

Under field conditions, tests were carried out in two phases: at the start of a new plan-tation and in a recently established one during two growth cycles. Fields, naturally infested with M. incognita and R. similis, respectively, were used. In the first phase, the in vitro plants were inoculated with the bioproduct of P. lilacinus, 20 days before planting. The nematode population on the roots was evaluated every two months using the maceration and sieving method. The bioproduct was applied around the plants (as a nematicide) on the recently established plantations (with 2000 R. similis/100 g of roots), and nematode population fluctuations were recorded every two months by the same method. The yield was evaluated over two crop cycles.

Bacillus thuringiensis (Bt)This biocontrol agent was also evaluated under various conditions. Different strains were screened under in vitro conditions against M. incognita using the same method described above. Under field conditions, the bioproduct containing the strain LBT-3 was applied on a large scale in more than 15 fields of different sizes. The product was deposited around the plant. The nematode populations were evaluated every two months by the same method.

Corynebacterium paurometabolumThis bacterium was evaluated under in vitro conditions, semi-controlled and field con-ditions using the same methodology.

Mycorrhyzae (Arbuscular Mycorrhizal Fungi, AMF)After a survey of natural populations of mycorrhyzae in banana and plantain fields, some species of the Glomus genus were selected to be evaluated under semi-controlled and natural conditions. The inoculum of mycorrhyzae was applied to in vitro plants at the hardening stage in both experiments. The nematode populations were evaluated with the same methodology and some growth indicators and yields were recorded.

Results and discussionP. lilacinus showed a positive effect, reducing the colonization and hatching of Meloidogyne egg masses (Table 1). R. similis treated with this fungus showed com-plete immobility, malformations of the digestive system and also vacuoles 48 hours

Table 1. Effect of P. lilacinus on M. incognita under in vitro conditions. Concentration Egg masses Hatching (%) Observations on second conidia/g colonized (%) stage juveniles (J2)

100 (Control) 0 71 J2 with motility

105 40 40 J2 with motility

106 40 45 J2 with motility

107 100 24 J2 with poor motility, growth of hyphae

108 100 5 J2 without movement

109 100 3 J2 without movement


after treatment. A toxin seems to be involved. P. lilacinus influenced the motility of nematodes in both cases and also affected the hatching of M. incognita eggs and the morphology of R. similis. This behaviour has important implications for root infec-tion by nematodes. Under semi-controlled conditions, the preventive treatment of P. lilacinus in hardened in vitro plants reduced the number of M. incognita nematodes from 8900 to 820 /50 g roots. According to these results, it is recommended that biolo-gical control begin before nematode infestation in order for the fungus to colonize the rhizosphere. This type of preventive treatment has shown good and practical results in banana production systems. The preventive use of P. lilacinus on in vitro plants showed encouraging results in plantations with low initial nematode populations. During one year, the populations of M. incognita and R. similis did not increase significantly with respect to the control (Table 2). Also under these conditions, yields were 25% higher in treated plants.

Table 2. The effect of P. lilacinus on the number of M. incognita and R. similis over 15 months under field conditions. P. lilacinus increased yield by 25%.Treatment Nematode Months

3 6 9 12 15

Control 460 990 2 500 4 700 8 900

M. incognita

P. lilacinus 150 320 450 818 1 560

Control 520 1 000 3 100 5 200 9 200

R. similis

P. lilacinus 100 250 410 790 1 700

Under production conditions, in recently established plantations with low initial nema-tode populations, the P. lilacinus treatments produced remarkable reductions in R. simi-lis and M. incognita, between 75-85%, over a period of more than one year. Therefore, this fungus was used on a commercial scale in more than 5500 ha. This fungus showed good results in nematode control in different crops. Segers and Butt (1994) reported the use of the bionematicide BIOSTAT (with P. lilacinus base) to control species of Meloidogyne, Pratylenchus, Scutellonema and Helicotylenchus while Holland (2001) reports the efficiency of another bio-pesticide with the same fungus on banana nemato-des. However, P. lilacinus is not currently being used in Cuba pending the completion of toxicology tests.

There are more than 80 strains in the Bt collection at the Instituto de Investigaciones de Sanidad Vegetal (INISAV). Of the 37 most-studied strains, 16 showed some nemati-cide activity. Strains LBT-1, LBT-3, LBT-4, LBT-25 and LBT-47 produced irreversible effects on egg hatching of M. incognita and have been selected for further study. LBT-3 strain is currently used to control some Lepidoptera and its large-scale reproduction methodology is well known. For this reason, it was selected for application under pro-duction conditions in the Camaguey province that suffers from significant nematode problems. An average nematode reduction of 87% was obtained two months after the treatments on more than 165 ha (Table 3). This strain is normally reproduced in


specialized centers (CREE) and applied under production conditions in Eastern provin-ces such as Camaguey. Our results are similar to those obtained by Zuckermann et al. (1993) and Sharma (1995), who reported different degrees of nematode control with Bt strains. Also, very efficient strains were mentioned in Current Drugs Patents (2002). The effects of Bt against nematodes are attributed to the activity of toxins produced by the sporulated bacteria (Rodriguez et al. 1991).

Studies on C. paurometabolum showed that it affects eggs and juveniles under labora-tory and field conditions. An almost total inhibition of M. incognita egg hatching (98%)

Table 3. Technical effectiveness (TE) of the B. thuringiensis var. kurstaki strain LBT-3 under production conditions in the Camaguey province, Cuba.Location Field (number) Area (ha) TE

Empresa Cultivos Varios Florida 53 4.3 100

(UBPC Celia Sánchez) 54 2.5 100

55 3.0 100

Empresa Pecuaria Céspedes 6 4.0 80

(UBPC 10 de Octubre) 7 3.4 100

9 9.4 100

11 6.7 91

12 9.4 90

14 3.4 100

15 3.4 88

18 5.4 100

(Distrito Angel del Castillo) 31 5.0 64

Empresa Pecuaria Esmeralda 34 9.0 100

(UBPC Lidia and Clodomira) 41 10.7 80

Empresa Cultivos Varios Camalote 18 9.1 84

(Distrito Manuel Boza) 19 9.1 57

20 4.3 62

21 5.4 100

22 3.4 83

23 3.0 94

24 6.3 70

26 5.3 84

(Distrito Las Flores) 2 4.2 80

6 3.7 80

Empresa Cultivos Varios Camagüey 11-B 4.7 92

(UDCT Victoria 1) 4 4.0 97

1 5.9 91

2-B 5.4 100

(UBPC Aguacate) 28 6.7 94

(UBPC 1ro de Enero) 3 5.4 100

TOTAL 165.4 87


was observed in the former. Its mode of action is due mainly to the combined effect of the culture media where the bacteria grew (chitinases) and the gases it produced (sulphides). Testing under controlled con-ditions with R. similis indicated an almost 85% reduction in the population compared with the control (Table 4). According to the mode of action, the gases and metabolites could penetrate inside the roots through cavities formed by nematodes. Under field conditions, it was noticed that R. similis decreased after the treatments with the bacte-ria or phenamiphos. While this behaviour was maintained throughout the experiment (12 months) in the fields treated with bacteria, populations in the phenamiphos treated fields started to increase at the seventh month (Figure1). The yields of treated plants were significantly higher than those of the control plants, with increases of 106% for

Figure 1. Behaviour of R. similis populations in plantain ‘Macho 3/4’ over time from planting. Right hand axis is yield, expressed as percentage increase with respect to control.

Table 4. Effectiveness of C. paurometa-bolum and B. thuringiensis on R. similis under controlled conditions. Treatment Nematodes/plant

C. paurometabolum 100 ml 5 b

B .thuringiensis 100 ml 50 b

Control 2345 aMeans of nematodes/plant followed by the same letter are not different (P=0.05) according to Duncan multiple range test.

200

180

160

140

120

100

80

60

40

20

0

Month

75

50

712

5

47

1021

12.5.

180

30

125

200

160

175 120

100

80

60

40

20

0

106

66

Yields(percentage increase

with respect to control)

7 9 1252

C. paurometabolum

Phenaphibos

Control

Nematode population/sample(Percentage of initial population)

% %

the bacterium and 66% for phenamiphos. There are no reports in the literature of this bacterium acting as a nematicide. Currently, a large scale reproduction methodology is available and a bio-nematicide product is in the registration and patent phase.

Regarding mycorrhyzae, early inoculation of in vitro plants during the hardening stage with Glomus species attenuated damage caused by R. similis and M. incognita and also reduced the nematode population. Testing under controlled conditions demonstrated that populations of both species were reduced to different degrees, with G. intraradices,


G. manihotis and G. mosseae the most promising. Field tests in plantations with low initial infestation have shown that plants inoculated early with AMF developed better than non-inoculated ones and that nematode populations grew more slowly than in control plots. Also, in one test using the FHIA-18 variety (resistant to R. similis), there was a tendency in plants with AMF to have better plant growth indicators and yields (Table 5). Jaizme-Vega (2001) and Sarah (2001) have reported similar results with banana nematodes and different arbuscular mycorrhizal fungi. These researchers sug-gest the possibility of including this alternative in management systems.

Table 5. Effect of colonization with mycorrhizae on plant growth and yields of the FHIA-18 variety in the Villa Clara province. Treatment Plant height Circumference Number of Bunch weight (m) (cm) leaves (kg)

Mycorrhizae 2.18 41.8 7.33 17.1

No mycorrhizae 2.14 40.0 6.33 14.1

Significance NS NS NS NSNS = no significant difference between means.

ConclusionsBiological control of plant parasitic nematodes in banana and plantain showed great potential for use under field conditions. However, it is advisable to perform local tests to study the necessary adaptations according to the different agro-ecosystems. P. lilaci-nus is an effective biocontrol agent of R. similis and M. incognita. The best results were achieved with the preventive inoculation of healthy in vitro plants during the hardening phase.

The strain LBT-3 of B. thuringiensis has nematicidal activity on R. similis and M. inco-gnita under field conditions. Nevertheless, other strains have shown promising effects and will be studied in the near future. C. paurometabolum has a good nematicidal acti-vity on the main banana and plantain parasitic nematodes. Its effectiveness has been demonstrated under a range of conditions. Mycorrhyzae of the genus Glomus reduced nematode damage caused by R. similis and M. incognita. The species G. intraradices, G. manihotis and G. mosseae have shown the most stable results.

ReferencesCurrent Drug Patents. 2002. (On line) www.web.com 2002. Consulted 9 May 2002.

Holland R. 1998. PAECIL (On line) www.ticorp.com.au/article 1 htm1998 Consulted 12 November 2001.

Jaizme-Vega M.C. 2001. Individual partner annual reports ICIA-IRTA. in Fourth Annual Report November 2000-October 2001. Project INCO. No. ERB IC18 CT 97-0208. 111pp.

Rodríguez M., M. de la Torre & E. Urquijo. 1991. Bacillus thuringiensis: Caracteristicas biológicas y perspectivas de producción. Revista Latinoamericana de Microbiología 33:280.

Sarah J.L. 2001. Individual partner annual reports CIRAD. in Fourth Annual Report November 2000-October 2001. Project INCO. No. ERB IC18 CT 97-0208. 111pp.

Segers R. & T.M. Butt. 1994. The nematophagus fungus Verticillium chlamydosporium produces a chymoelastase-like protease which hydrolyses host nematode proteins in situ. Microbiology 140:2715-2723.

The problem of banana root deterioration and its impact on production200 Biological control of nematodes in banana C.A. Gauggel et al. 201M. Araya and T. Moens

Sharma R.D. 1995. Bacillus thuringiensis: a biocontrol agent of plant parasitic nematodes. Pp. 251-254 in Congreso Internacional de Nematologia Tropical. Program and Proceedings:. Sociedad Brazileira de Nematologia, Brazil.

Taylor A.L. & J.N. Sasser. 1978. Biology, Identification and Control of Root-knot nematodes. A Cooperative Publication of Department of Plant Pathology North Carolina State University and USAID. 111pp.

Zuckermann B.M, M.B. Dicklow & N. Marban Mendoza. 1993. Nematocidal Bacillus thuringiensis biopesticides. PCTWO 93/19604.Tropicales (INIVIT). Villa Clara. Cuba [email protected]

The problem of banana root deterioration and its impact on production200 Biological control of nematodes in banana C.A. Gauggel et al. 201M. Araya and T. Moens

Parasitic nematodes on Musa AAA (Cavendish subgroup cvs ‘Grande naine’, ‘Valery’ and ‘Williams’)M. Araya1 and T. Moens2

AbstractAmong the biotic factors affecting yield, banana-root nematodes are second only to black Sigatoka (Mycosphaerella fijiensis). The three commercial cultivars, ‘Grande naine’, ‘Valery’ and ‘Williams’ are equally susceptible, and in plantations in Costa Rica usually only polyspecific communities occur, consisting of a mixture of the migratory endoparasitic Radopholus similis and Pratylenchus coffeae, the ecto-endoparasitic Helicotylenchus multicinctus, and the sedentary endoparasitic Meloidogyne incognita, or rarely M. javanica. Based on their frequencies and population densities the relative importance of nematode genera in Costa Rica was established in decreasing order as follows: R. similis > Helicotylenchus spp. > Meloidogyne spp. > Pratylenchus spp. R. similis is the most abundant nematode, accounting for 82 to 97% of the overall root population. High populations of R. similis are found throughout the year and in all counties producing bananas. The nematodes damage the root and corm tissue. R. similis enters the root mainly by the root tip, but any portion of the entire root length may be invaded. Adults and juveniles occupy an intercellular position in the cortical parenchyma, in which they move actively, causing damage as they feed on the cytoplasm of the surrounding cells. The four nematode genera develop and complete their life cycle inside the banana roots. In highly infected roots, R. similis sometimes crosses the endodermis and invades the stele. Reddish brown lesions appear throughout the cortex. In some Costa Rican banana growing areas, crop losses on nematode infected plantations can be high, up to 30-50%. Infected plants have poor root anchorage and the ability of the root system to take up water and nutrients is reduced, which results in losses in bunch weight and crop longevity, and a lengthening in the plant production cycle. Pathogenicity studies have shown that R. similis restricts banana root system growth and reduces the concentration of K in the root dry matter. All the phenological stages can be infected by any of the four genera, but again, R. similis is the most frequent and the most abundant at any stage. Even the roots of very small peepers (suckers 10 cm high) can be infected, and in the case of R. similis, it is common to find nematodes in the corm tissue also. This suggests that root infection of these young suckers in infected plants is caused by either R. similis coming from the corm or from the soil.

Resumen - Nematodos parásitos de Musa AAA (subgrupo Cavendish cvs ‘Grande naine’, ‘Valery’ y ‘Williams’)Entre los factores bióticos que afectan el rendimiento en banano, los nematodos son el segundo en importancia después de la Sigatoka negra (Mycosphaerella fijiensis). Los tres cultivares comerciales ‘Grande naine’, ‘Valery’ y ‘Williams’, son igualmente susceptibles y en las plantaciones locales, usualmente solo comunidades poliespecíficas ocurren, consistiendo de una mezcla de los endoparásitos migratorios Radopholus similis y Pratylenchus coffeae, el ecto-endoparásito Helicotylenchus multicinctus y el endoparásito sedentario Meloidogyne incognita o raramente M. javanica. Basado en las frecuencias y densidades poblacionales, la importancia relativa de los géneros de nematodos en Costa Rica se estableció en orden decreciente como sigue: R. similis > Helicotylenchus spp. > Meloidogyne spp. > Pratylenchus spp. R. similis es el nematodo más abundante y representa de un 82-97% de la población total en las raíces. Altas poblaciones de R. similis se encuentran durante todo el año en todos los cantones productores de banano. El daño

1 Corporación Bananera Nacional (CORBANA S.A.) Apdo 390, 7210 Guápiles, Costa Rica. e-mail: [email protected]. 2INIBAP, CATIE 7170 Turrialba, Costa Rica. e-mail: [email protected].

The problem of banana root deterioration and its impact on production202 Parasitic nematodes on Musa AAA C.A. Gauggel et al. 203M. Araya and T. Moens

se localiza en las raíces y cormo. R. similis penetra las raíces principalmente por la parte terminal (caliptra), pero cualquier parte de la raíz puede ser invadida. Adultos y juveniles ocupan una posición intercelular en el parénquima cortical donde ellos se mueven activamente causando daño conforme se alimentan del citoplasma de las células vecinas. Los cuatro géneros de nematodos se desarrollan y completan su ciclo de vida dentro de las raíces de banano. En raíces altamente infectadas, R. similis algunas veces pasa la endodermis e invade el cilindro vascular. Coloraciones café rojizas aparecen a través de toda la corteza. En algunas plantaciones locales infectadas las pérdidas en rendimiento alcanzan hasta un 30-50%. Plantas infectadas carecen de buen anclaje y la habilidad de las raíces para absorber agua y nutrientes se reduce, lo que resulta en pérdida de peso del racimo y longevidad de la planta y se alarga los intervalos entre ciclos de producción. Estudios de patogenicidad muestran que R. similis restringe el crecimiento del sistema radical y reduce la concentración de K en las raíces. Todos los estados fenológicos de la planta de banano pueden ser infectados por cualquiera de los cuatro géneros, pero nuevamente R. similis es el más frecuente y abundante en cualquier estado de desarrollo de la planta. Aún raíces de hijos muy pequeños (10 cm de altura) pueden ser infectadas y en el caso de R. similis es común también encontrarlo en el cormo. Esto sugiere que la infección de las raíces en hijos nuevos de plantas infectadas, ocurre igualmente de R. similis procedente del cormo o del suelo.

IntroductionBanana is an important crop in Costa Rica accounting for almost 17% of the agricul-tural gross national product. In 2003, 1.85 million tonnes were exported, produced on 42,000 ha, generating a total income of US $541 million FOB (Sánchez and Zuñiga 2004). Besides the constraints of banana market requirements and demands, there are other limiting factors. Among the biotic factors affecting yield, banana-root nematodes are second only to black Sigatoka (Mycosphaerella fijiensis). Nematodes reduce bunch weight and plant longevity, and increase the crop cycle duration (Quénéhervé 1991).

In Costa Rican plantations (Araya et al. 1995, 2002) usually only polyspecific com-munities occur, consisting of a mixture of Radopholus similis, Helicotylenchus multi-cinctus, Meloidogyne incognita, M. javanica, and Pratylenchus spp. Here we report some research results related to the parasitic behaviour and nematode damage under controlled conditions on commercial banana plantations in Costa Rica.

Host suitabilityThe commercial cultivars planted are ‘ Grande naine’, ‘Valery’ and ‘Williams’, which are part of the Musa AAA Cavendish subgroup. All these cultivars are equally suscep-tible to R. similis (Figure 1) and their reaction to each of the other three most important

11596 12386 11092

Valery Grande naineCultivar

R.similis

/10

0gof

roots

Williams

15000

12000

9000

6000

3000

0Figure 1. Radopholus similis numbers in banana (Musa AAA subgroup Cavendish cvs ‘Valery’, ‘ Grande naine’

and ‘Williams’). Each bar is the mean ± standard error of 250 root samples. Each sample contains roots

of 5 plants.


Musa parasitic nematodes is similar. This reaction agrees with pot experiments, carried out under our conditions with R. similis (Moens et al. 2003) and with the other nema-todes (Araya et al. 2003). Also, other researchers found similar results, inoculating Cavendish subgroup cultivars with R. similis (Stoffelen et al. 2000, Viaene et al. 2000 and 2003), M. incognita (Pinochet et al. 1998, Van den Bergh et al. 2002a, b), P. coffeae (Viaene et al. 1998, Stoffelen et al. 1999a, 2000, Viaene et al. 2000, Van den Bergh et al. 2002a) and H. multicintus (Barekye et al. 2000).

Nematode frequencies and populations In surveys in Costa Rica from 1994 to 1999, looking at root samples extracted from five plants on each occasion, R. similis was the most abundant nematode, ranging from 82 to 86%, while Helicotylenchus spp., Meloidogyne spp. and Pratylenchus spp. varied respectively from 8 to 10%; 5 to 9% and < 1% of the root population (Araya et al. 1995, 2002). The four nematode genera detected are well known pathogens in banana roots (McSorley and Parrado 1986, Sarah 1989, Gowen 1995, Sarah et al. 1996, Bridge et al. 1997, De Waele and Davide 1998, Marin et al. 1998).

Nematode frequency was very stable in the different years (Table 1). The highest frequency, above 95%, was always found for R. similis, followed by Helicotylenchus spp., ranging from 52 to 60%. The four nematode genera counted were detected in all counties and all months of the year. The favorable climatic conditions (Figure 2), without a dry season and only small variations in temperature, promote adequate plant growth throughout the year. This may explain the low variation observed in nematode population densities and frequencies.

Table 1. Absolute frequency by year (1994-1999) of the banana (Musa AAA) root parasitic nematodes in Costa Rica as a percentage of the total number of samples. Year

94 95 96 97 98 99 Total

No. of samples 11 688 10 618 11 253 10 929 12 707 14 525 71 720

Percentage of samples with nematode species Mean

R. similis 95 96 96 99 98 97 97

Helicotylenchus 56 56 52 54 56 60 56

Meloidogyne 68 59 51 57 48 47 56

Pratylenchus 15 13 11 13 13 14 13

The distribution of the root samples per nematode population density interval clearly indicated that R. similis accounted for the highest population (Figure 3). From the 71 720 recorded root samples, only 2324 (3%) were free of the nematode and 37 489 (52%) contained more than 10 000 individuals per 100 g of roots. These high popula-tions and frequencies were probably encouraged by the long-term banana monoculture and the affinity between this nematode and the type of commercial banana planted (Musa AAA) (Orton and Siddiqi 1973). The high populations of R. similis agree with its high reproductive fitness in vitro on carrot discs (Fallas and Sarah 1995, Stoffelen et


al. 1999b) and on banana plants under controlled conditions (Fallas et al. 1995, Binks and Gowen 1997, Stoffelen et al. 1999c).

The high R. similis populations and frequencies are consistent with other local studies (Jiménez 1972, López 1980, Figueroa 1989) and with observations from the Philippines (Boncato and Davide 1992), Colombia (Gómez 1980), and Ecuador (Gómez 1997). The behavior of R. similis as a main banana root nematode agrees with the results of Fargette and Quénéhervé (1988), Speijer and Fogain (1999), and Kashaija et al. (1999) in African countries, Stanton (1994) in Australia, Davide (1994) in the Philippines, Pone (1994) in the Pacific Islands, Jiménez et al. (1998) in Ecuador and Gómez (1980) in Colombia.

Helicotylenchus spp. was absent in 31,608 (44%) of the samples and in only 1,779 (2%) reached levels above 10 000 nematodes per 100 g of roots (Figure 3). Helicotylenchus spp. populations were lower than R. similis, in agreement with other results (Gómez 1980). More likely, this is because this nematode does not reproduce as fast as R. similis and the life cycles of the nematodes differ. H. multicinctus was found to have a life cycle of 42 days at 28oC on Arabidopsis thaliana and the adult females laid 4 eggs per day for a period of 10-12 days (Orion and Bar-Eyal 1995). In contrast, the R. similis life cycle was completed in 20-25 days at 24-32oC on banana roots and the adult females laid 4-5 eggs per day for 15 days (Loos 1962). This means that R. similis could be expected to have more generations and more individuals per generation in the same period of time.

More than 33,564 (46%) samples were free of Meloidogyne spp. and only 1113 (2%) showed densities higher than 10,000 nematodes per 100 g of roots (Figure 3). This low population density and frequency could be related to the feeding behavior of R. similis. Pinochet (1977) and Santor and Davide (1992) found that the presence of R. similis on the galls, induced by M. incognita, caused deterioration and disintegration of the giant cells, which, in turn, affected the development and reproduction of M. incognita.

0

5

10

15

20

25

30

35

J F M A M J J A S O N DMonth

Tempe

rature

�C

0

100

200

300

400

500

600

Rainfall(m

m)

Rainfall Mean T�Maximum T� Minimum T�

Figure 2. Average of 15 years (1989-2003) of the monthly mean, maximum and minimum temperature (˚C) and rainfall (mm) in a Costa Rican export banana-growing area.


2417 (3,37%)

1364 (1,90%)

21014 (29,30%)

251 (0,35%)

325 (1,45%)

16012 (22,33%)

21825 (30,43%)

850 (1,19%)

20113 (28,04%)

14724 (20,53%)

164 (0,23%)

2370 (3,30%)

3823 (5,33%)

15613 (21,77%)

6555 (9,14%)

1205 (1,68%)

6808 (9,49%)

8080 (11,27%)

8248 (11,50%)

4762 (6,64%)

8130 (11,33%)

27865 (38,85%)

26430 (36,85%)

8046 (11,22%)

423 (0,59%)

62172 (86,68%)

33564 (46,80%)

31608 (44,07%)

2324 (3,24%)

1444 (2,01%)

45 (0,07%)

0 15000 30000 45000 60000 75000Frequency

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

0 (0,00%)

4 (0,01%)

12 (0,01%)

10 (0,01%)

Radopholu

ssimilis

Helico

tyle

nch

ussp

p.

Melo

idogy

ne

spp.

Pra

tyle

nch

ussp

p.

Tota

lnemato

des

Nemato

desper100

gofro

ots

Figure 3. Frequency of nematodes according to the specific ratios per 100 g of fresh roots in 71 720 banana (Musa AAA) root samples recorded from 1994 to 1999. Total nematodes = R. similis + Helicotylenchus spp. + Meloidogyne spp. + Pratylenchus spp.


Pratylenchus spp. were only present in 9548 (13%) samples, with just 49 (0.08%) above 10,000 nematodes per 100 g of roots (Figure 3). This is reasonable, because Pratylenchus spp. has the same habitat as R. similis and a longer life cycle (Siddiqi 1972). This means that while they compete for the same feeding site, R. similis, with its higher reproduction rate, suppresses the growth of Pratylenchus spp.

When all nematodes were pooled (total nematodes), only 423 samples (1%) were nema-tode free and 45,256 (63%) samples contained more than 10,000 nematodes per 100 g of roots (Figure 3). Similar frequency, abundances and distributions were observed when sampling was done at flowering and harvest of individual plants on three expe-riments (Figures 4, 5 and 6). Only in the experiment carried out on the newest farm (Figure 6), which was planted about nine years ago, produced samples free (4%) of any parasitic nematode (Araya et al. 2003).

Nematode populations according to sucker height and phenological stage of the plantThe suitability of different phenological stages of the plant for parasitic nematodes was evaluated in five experiments by sampling roots of suckers of different heights and by sampling plants at flowering, harvest and 30, 60 and 90 days after harvest (Araya et al. 2003). On three farms, Anabel, Calinda, and Esmeralda, no difference in the R. similis population was found among suckers from 10 to 350 cm height or even with recently flowered or harvested plants (Figure 7). The lowest R. similis population found on any of the farms and at any sucker height or plant stage, was higher than 14,000 nematodes per 100 g of roots.

Helicotylenchus spp. populations were low with the exceptions of those found on the Anabel and Nueva Esperanza farms. On Anabel, the highest populations were observed in suckers of 340-350 cm height, while on Nueva Esperanza, populations increased up to the flowering stage (Figure 8). When individual plants were sampled five times progressively (at flowering, at harvest, and 30, 60 and 90 days after harvesting), the R. similis population decreased as plants aged on the three farms. The lowest popu-lation was consistently found to be at 90 days after harvest (Figure 9). Nevertheless, the nematodes that remain in the harvested plants act as a reservoir and can become a source of inoculum for the other members of the plant production unit.

Pathogenicity and symptoms of Radopholus similisMoens et al. (2003) found a linear reduction of root weight when R. similis was inoculated at densities increasing from 0.14 to 2.24 nematodes per mL of substrate, corresponding to 254 to 2128 R. similis per pot, respectively. Root weight decreased by 3.9 g (16%) with each successive inoculation of 1000 nematodes (Figure 10A). Sarah et al. (1993) found that after exposing ‘Valery’ plants in 1.8 L pots to R. similis populations from different places around the world, root weights decreased from 19 to 80%, 8 weeks after inoculation with 300 individuals. Also, Marin et al. (1999), using ‘Grande naine’ plants in 1.4 L pots inoculated with 200 R. similis from different coun-tries, observed root weight reductions of up to 21% after 8 weeks. Hahn et al. (1996) found decreases in root weight ranging from 8 to 26% in ‘Valery’ plants in 0.8 L pots 8 weeks after inoculation with 100 R. similis, and up to 30% 12 weeks after inoculation


01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

2 (1,94%)

15 (14,56%)35 (33,98%)

43 (41,75%)

8 (7,77%)

0 (0,00%)

2 (2,00%)

25 (25,00%)

25 (25,00%)

9 (9,00%)

1 (1,00%)

0 (0,00%)

38 (38,00%)

0 (0,00%)

3 (3,00%)

11 (11,00%)29 (29,00%)

42 (42,00%)

14 (14,00%)

1 (1,00%)

4 (3,88%)

24 (23,30%)

30 (29,13%)20 (19,42%)

2 (1,94%)0 (0,00%)

23 (22,33%)

81 (78,64%)

2 (1,94%)0 (0,00%)0 (0,00%)

0 (0,00%)

0 (0,00%)

20 (19,42%)

71 (68,93%)

1 (0,97%)0 (0,00%)0 (0,00%)

0 (0,00%)

0 (0,00%)

0 (0,00%)0 (0,00%)

0 (0,00%)

0 (0,00%)

31 (30,10%)

77 (77,00%)

0 (0,00%)

0 (0,00%)0 (0,00%)

0 (0,00%)

0 (0,00%)

23 (23,00%)

58 (58,00%)

3 (3,00%)39 (39,00%)

0 (0,00%)

0 (0,00%)

4 (3,88%)

24 (23,30%)

60 (58,25%)

15 (14,56%)

0 (0,00%)0 (0,00%)

1 (1,00%)

6 (6,00%)

24 (24,00%)

51 (51,00%)

18 (18,00%)

0 (0,00%)

0 (0,00%)

Radopholu

ssimilis

Helico

tyle

nch

ussp

p.

Melo

idogy

ne

spp.

Nemato

desper100

gofro

ots

Pra

tyle

nch

ussp

p.

Tota

lnemato

des

0 30 60 90 120 150Frequency Frequency

0 30 60 90 120 150

Figure 4. Frequency of nematodes according to the specific ratios in 103 individual banana (Musa AAA) root samples recorded on the Productora Tropical farm. Total nematodes = R. similis + Helicotylenchus spp. + Meloidogyne spp.+ Pratylenchus spp.


01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

0 (0,00%) 1 (1,04%)7 (6,67%) 7 (7,29%)5 (4,76%) 8 (8,33%)

27 (25,71%) 14 (14,58%)31 (29,52%) 31 (32,29%)29 (27,62%) 31 (32,29%)

6 (5,71%) 4 (4,17%)

7 (6,67%) 7 (7,29%)39 (37,14%) 36 (37,50%)

30 (28,57%) 17 (17,71%)18 (17,14%) 28 (29,17%)

10 (9,52%) 6 (6,25%)1 (0,95%) 1 (1,04%)0 (0,00%) 1 (1,04%)

74 (70,48%) 75 (78,13%)30 (28,57%) 20 (20,83%)

1 (0,95%) 0 (0,00%)0 (0,00%) 1 (1,04%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)

91 (86,67%) 87 (90,63%)14 (13,33%) 9 (9,38%)

0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)

0 (0,00%) 0 (0,00%)1 (0,95%) 1 (1,04%)3 (2,86%) 5 (5,21%)

19 (18,10%) 14 (14,58%)30 (28,57%) 27 (28,13%)

45 (42,86%) 43 (44,79%)7 (6,67%) 6 (6,25%)

Radopholu

ssimilis

Helico

tyle

nch

ussp

p.

Nemato

desper100

gofro

ots

Melo

idogy

ne

spp.

Pra

tyle

nch

ussp

p.

Tota

lnamato

des


0 30 60 90 120 150

Figure 5. Frequency of nematodes according to the specific ratios in 105 individual banana (Musa AAA) root sam-ples recorded on the San Pablo farm. Total nematodes = R. similis + Helicotylenchus spp. + Meloidogyne spp. + Pratylenchus spp.


01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

01 a 2500

2501 a 50005001 a 10000

10001 a 2000020001 a 50000

> 50000

22 (19,13%) 20 (17,54%)20 (17,39%) 24 (21,05%)11 (9,57%) 7 (6,14%)15 (13,04%) 15 (13,16%)18 (15,65%) 17 (14,91%)

22 (19,13%) 28 (24,56%)7 (6,09%) 3 (2,63%)

100 (86,96%) 103 (90,35%)

13 (11,33%) 10 (8,77%)2 (1,74%) 0 (0,00%)0 (0,00%) 1 (0,88%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)

42 (36,52%) 42 (36,84%)60 (52,17%) 53 (46,49%)

4 (3,48%) 10 (8,77%)5 (4,35%) 6 (5,26%)2 (1,74%) 2 (1,75%)2 (1,74%) 1 (0,88%)0 (0,00%) 0 (0,00%)

114 (99,13%) 114 (100,00%)

1 (0,87%) 0 (0,00%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)0 (0,00%) 0 (0,00%)

4 (3,48%) 0 (1,75%)

26 (22,61%) 30 (26,32%)14 (12,17%) 10 (8,77%)20 (17,39%) 17 (14,91%)20 (17,39%) 23 (20,18%)24 (20,87%) 29 (25,44%)

7 (6,09%) 3 (2,63%)

Radopholu

ssimilis

Helico

tyle

nch

ussp

p.

Melo

idogy

ne

spp.

Pra

tyle

nch

ussp

p.

Tota

lnamato

des


0 30 60 90 120 150

Nemato

desper100

gofro

ots

Figure 6. Frequency of nematodes according to the specific ratios in 115 individual banana (Musa AAA) root sam-ples recorded on the La Rebusca farm. Total nematodes = R. similis + Helicotylenchus spp. + Meloidogyne spp. + Pratylenchus spp.


05000

1000015000200002500030000350004000045000

010000200003000040000500006000070000

05000

1000015000200002500030000350004000045000

0150003000045000600007500090000

105000

010000200003000040000500006000070000

Rado

pholus

similispe

r100gof

roots

Anabel

Calinda

Esmeralda

Nueva Esperanza

RebuscaP = 0.0002

P = 0.0043

P = 0.3571

P = 0.1271

P = 0.1309

10cm

40-50cm

90-100cm

140-150cm

190-200cm

240-250cm

290-300cm

340-350cm

Flowering

Harvest

Sucker height (cm) and phenological plant stage

Figure 7. Radopholus similis according to sucker height (cm) or phenological banana (Musa AAA cv. ‘Grande naine’) plant stage. Each point is the mean ± standard error of 15 samples. All the samples were taken in a maximum of 3 days. Each sam-ple was composed of the roots of 5 plants. The p values correspond to the comparison among the means of the diffe-rent sucker heights or plant stages. The names in the upper right corner are the farms where the experiment was done.


Figure 8. Helicotylenchus spp. according to sucker height (cm) or phenological banana (Musa AAA cv. ‘Grande naine’) plant stage. Each point is the mean ± standard error of 15 samples. All the samples were taken in a maximum of 3 days. Each sample was composed of the roots of 5 plants.

30000

25000

20000

15000

10000

5000

0

6000

5000

4000

3000

2000

1000

0

1400

1200

1000

800

600

400

200

0

52500

45000

37500

30000

22500

15000

7500

0

27002400210018001500900600300

0

Helicotylen

cuspe

r100gof

roots

Anabel

Calinda

Esmeralda

Nueva Esperanza

RebuscaP = 0.3515

P = 0.0001

P = 0.3302

P = 0.1394

P = 0.3588

40-50cm

90-100cm

140-150cm

190-200cm

240-250cm

290-300cm

340-350cm

Flowering

Harvest

Sucker height (cm) and phenological plant stage


with 300 R. similis. Fallas et al. (1995), evaluating ‘Valery’ plants in 0.8 L pots inocu-lated with 100 R. similis, found decreases in root weight of 11 to 53% after 12 weeks. A reduction in the K concentration in the root dry matter was observed when ‘Grande naine’ plants were inoculated with different initial R. similis populations (Figure 10B). Other nutrients that varied significantly were N, S, Ca and Cu.

Fogain and Gowen (1995) found root necrosis of 3 to 12%, 6 weeks after inoculating ‘ Grande naine’ plants in 1 L pots with 300 R. similis. Fogain et al. (1996) observed

Productora Tropical (N = 103)

0

5000

10000

15000

20000

25000

30000

35000 Rs He Me Pr

Rebusca (N = 115)

0

5000

10000

15000

20000

25000

30000

35000

San Pablo (N= 105)

0

5000

10000

15000

20000

25000

30000

35000

Nem

atod

espe

r100gof

roots

Flowering

Harvest

30daysafter harvest

60daysafter harvest

90daysafter harvest

Sampling times

Figure 9. Nematode numbers in individual plants sampled 5 times progressively (at flowering, at harvest, and 30, 60 and 90 days after harvest). Rs = Radopholus similis, He= Helicotylenchus spp., Me = Meloidogyne spp. and Pr = Pratylenchus spp.


a high root lesion index in follower suckers of plantain cultivars in 10 L pots inocula-ted with 1000 R. similis. Moens et al. (2001) found a significant correlation between R. similis and root necrosis or damage, ranging from 0.62 to 0.75, in root samples from commercial banana plantations.

‘Giant Cavendish’ bananas inoculated with 1000, 2000, 3000 or 4000 R. similis suffered bunch weight reductions of 17, 42, 51, and 61% respectively (Davide and Marasigan, 1985). In a local study, ‘ Grande naine’ plants, initially grown in 1.8 L pots, were inocu-lated with 1000 R. similis or M. incognita or P. coffeae or H. multicinctus. The R. similis and P. coffeae were cultured on carrot disks and M. incognita and H. multicinctus on potted banana plants. Three weeks later, plants were transferred to 200 L drums con-taining sterilized soil and allowed to develop until harvest, when root content, damage

y = 7,0117x + 23,115R2 = 0,9275P = 0,006

0

5

10

15

20

25

Root

weight(g)

%Kof

root

dryweight

y = 0,0002x + 4,0943R2 = 0,38P = 0,0026

0

1

2

3

4

5

0 500 1000 1500 2000 2500 3000 3500 4000

Inoculated Radopholus similis densities or numbers

A

B

0 0.14 0.28 0.56 1.12 2.24

Figure 10. Effect of initial Radopholus similis inoculation densities per ml of substrate or numbers per 1.8 L pots on Musa AAA cv. Grande naine. A) root weight. Each point is the mean ± standard error of 12 repetitions over 8 weeks of exposure. B) % K of the dry root weight. Each point is the mean ± standard error of 12 repetitions, with an exposure time of 12 weeks.


and bunch weight were recorded. Plants inoculated with R. similis had 3.6 kg (66%) less root weight, 6.4 kg (27%) less bunch weight and 262% more root damage than non-inoculated plants (Figure 11A-B).

The root and corm symptoms (Figure 12A-D) agree with those reported in the literature (Tarté and Pinochet 1981, Gowen and Quénéhervé 1990, Gowen 1995, Sarah 2000). Black and brown discoloration was observed in roots and corm after washing the sur-face free of soil. However, in early infections pre-symptomatic roots (Figure 12C) hold a lot of nematodes and when there is excess soil water, symptoms are easily confused with those caused by waterlogging. When the corm was lightly peeled, a purple, red-dish-brown discoloration was easily seen, similar to that found in longitudinally split roots, where lesions extended from the epidermis towards the vascular cylinder.

Pathogenicity and symptoms of Meloidogyne incognitaIn pot experiments with 22 Vietnamese banana genotypes, fresh root weight increased significantly in 2 of the 3 experiments in plants inoculated with 500 to 4000 Meloidogyne

0

1

2

3

4

5

6

7

8

9

Root weight Root damage

Root

weigh

t(kg)

andda

mag

e

Helicotylenchus multicinctus

Helicotylenchusmulticinctus

Meloidogyneincognita

Pratylenchuscoffeae

Radopholussimilis

Non-inoculatedplants

Meloidogyne incognita

Pratylenchus coffeae Radopholus similis

Non-inoculated plants

A

B

0

5

10

15

20

25

30

Bunc

hweigh

t(kg)

Figure 11. Root fresh weight, root damage and bunch weight at harvest of Musa AAA cv. ‘Grande naine’ plants inocula-ted with 1000 nematodes compared with plants without nematode inoculation. Each bar is the mean ± standard error of 10 observations. The scale used for root damage was from 0 to 10.


spp. (Van den Bergh et al. 2002a). Also, in very young banana roots under in vitro conditions, M. incognita can penetrate the roots and cause damage (Coosemans et al. 1994). Jonathan and Rajendran (2000a) found a significant reduction in root weight for the higher initial inoculations in their study working with increasing inoculations from 0 to 10 000 individuals in pots of 105 kg soil capacity. In contrast, no negative effect on growth was observed in Musa AAB plants, inoculated with M. incognita and cultivated in 10 L pots in Ivory Coast (Adiko 1989).

In Malaysia, Razak (1994) reported stunted growth and small bunches due to the infection of M. incognita in ‘Pisang nangka’ and ‘Pisang mas’. Similar effects on plant growth have been reported in India (Patel et al. 1996) where a delay in flowering was also observed. Van den Bergh et al. (2000) reported that Meloidogyne spp. had the most negative effect on the bunch characteristics of banana plants in Vietnam. High

A B

D

C

Figure 12. Radopholus similis and infection-induced symptoms in banana roots and corm. A) Adult female of R. similis at 20x magnification. B) In the center a healthy root (cream-white color) and on either side, roots with purple, reddish-brown discoloration extending from the epidermis to the vascular cylinder, C) On the right, two apparently healthy, asymptomatic roots containing 19 240 R. similis per 100 g of roots. On the right, three roots completely damaged first by R. similis (88 560 / 100 g of roots) and later by waterlogging, and D) Reddish-brown discoloration on the corm tissue.

Meloidogyne spp. populations, up to 66,000 per 100 g of roots, have been reported in ‘ Grande naine’ vitroplants in field conditions in Cameroon (Fogain 1994).

‘Giant Cavendish’ bananas, inoculated with 1000, 10 000 or 20 000 M. incognita, suf-fered bunch weight reductions of 26, 45 and 57% respectively (Davide and Marasigan 1985). In our experiment, carried out in 200 L drums, M. incognita (Figure 13A-C) did not significantly reduce root weight, 0.5 kg (9%), and did not significantly increase root damage, but it did significantly suppress bunch weight by 7.5 kg (32%) (Figure 11A-B). This reduction in yield coincides with Jonathan and Rajendran’s results (2000b), which found bunch weight increases of 31% when controlling this nematode with nematicide. The observed root symptoms (Figure 13D) are close to those reported by Sikora and Schlösser (1973), Pinochet (1977), Tarté and Pinochet (1981), De Waele and Davide (1998) and De Waele (2000). Deformation, bifurcation, swellings and stunting on pri-mary and secondary roots are the most common symptoms where fine roots are rare. Sometimes gall formation is observed close to the root tip.

Figure 13. Meloidogyne incognita and its symptoms indu-ced by the infection on banana roots. A) Infective stage of M. incognita (20x magnification). B) Stages of M. inco-gnita extracted from banana roots (20x magnification). C) Female of M. incognita in a banana root (20x magni-fication) and D) Swelling and deformation of roots with gall formation close to the root tip.

A B

D

C

Parasitic nematodes on Musa AAA216

Pathogenicity and symptoms of Pratylenchus coffeae Van den Bergh et al. (2002a) observed a reduction in fresh root weight in 3 expe-riments with 24 Vietnamese genotypes of banana inoculated with 1,000 P. coffeae. Pinochet (1978), Tarté (1980), Rodríguez (1990), Bridge et al. (1997) also found that Pratylenchus spp. damaged banana roots and reduced yield. In our experiment, carried out in 200 L drums, P. coffeae (Figure 14A) did not significantly reduce root weight - 0.58 kg (11%), but it did significantly increase root damage by 129%, and significantly suppressed bunch weight by 5.6 kg (24%) (Figure 11A-B). Symptoms (Figure 14B) induced by P. coffeae infection were very similar to those caused by R. similis and coincided with those described by Pinochet (1978), Bridge et al. (1997) and Gowen (2000a). Purple or black necrosis of the epidermis and cortical root parenchyma with surface cracks occurred.

Figure 14. Pratylenchus coffeae and its symptoms induced by the infection on banana roots. A) Adult of P. coffeae (20x magnification). B) Purple and black necrosis extending from the root epidermis to the vascular cylinder.

Pathogenicity and symptoms of Helicotylenchus multicinctusHelicotylenchus multicinctus and H. dihystera (McSorley and Parrado 1986, Davide 1996, Mani and Al Hinai 1996, Chau et al. 1997) damage the banana root system and reduce yield by 19% (Speijer and Fogain 1999) to 34% (Reddy 1994). There are mixed reports on the effect of H. multicinctus in potted plants. Barekye et al. (1999) found a root weight reduction in some H. multinctus inoculated plants compared with plants without nematodes. In Uganda, a non-significant reduction in root weight was observed in Musa AAA-East Africa cultivar ‘Kisansa’ plants inoculated with 1000 H. multicintus (Barekye et al. 2000). In greenhouse experiments, we observed fresh root weight reduc-tions in Musa AAA ‘Grande naine’ plants inoculated with 515 H. multicinctus.

In Florida, H. multicinctus can cause severe root damage resulting in the toppling of mature plants (McSorley 1986, Gowen 1995). According to McSorley and Parrado (1986), this nematode is more important in subtropical banana production areas. In Lebanon, H. multicinctus is considered the most important banana root parasitic nema-tode (Sikora and Schlösser 1973). The same authors felt that this nematode, together

A B

C.A. Gauggel et al. 217M. Araya and T. Moens


with M. incognita, was associated with the general decay of the banana root system observed.

In our experiment carried out in 200 L drums, H. multicinctus (Figure 15A) did not significantly reduce either root weight, 0.42 kg (8%), or increase root damage, 19%, or suppress bunch weight, 0.9 kg (4%) (Figure 11A-B). However, in Israel, bunch weight increased by 18% when controlling this nematode with nematicide (Minz et al. 1960). The symptoms (Figure 15B) in the plants infected with H. multicinctus in drums coincide with those observed by McSorley (1986) and Gowen (2000b). Tertiary roots appeared necrotic and fell off when the roots were handled. Thicker and larger roots showed small black-brown surface lesions and larger necrotic areas.

A B

Figure 15. Helicotylenchus multicinctus and its symptoms induced by the infection on banana roots. A) Adult of H. mul-ticinctus (20x magnification). B) Small black-brown root surface lesions and larger necrotic areas.

Practical implications of mixed nematode populationsThe different parasitic habits of the nematode genera present - migratory endoparasites (R. similis and Pratylenchus spp.), sedentary endoparasites (Meloidogyne spp.) and ecto-endoparasites, feeding on subsurface tissue (H. multicinctus) - are likely to exacer-bate root damage, because of lesions at feeding sites in the root cortex and throughout the root tissue. Usually control is recommended when R. similis population density exceeds a specific number per 100 g of roots, but the other nematodes also reduce the root system and yield. Therefore, development of nematode management tactics requires consideration of the damage caused by the total phytonematode population.

In current breeding programmes, resistance to black Sigatoka is, together with a high yield and good postharvest qualities, the most important factor in the screening pro-cess. Nevertheless, some programs could also incorporate resistance to R. similis. This is the case in different FHIA hybrids. Part of the ‘Pisang jari buaya’ group (Musa AA) demonstrated resistance (Wehunt et al. 1978, Viaene et al. 2000 and 2003), and this trait was, through the SH-3142 improved diploid (Pinochet and Rowe 1979), successfully incorporated into a series of FHIA hybrids. Other potential sources of resistance to R. similis are ‘Kunnan’ and ‘Paka’ (Musa AA section Eumusa) (Collingborn and Gowen 1997), and the Fe’i varieties ‘Menei’ and ‘Rimina’ (section Australimusa) (Stoffelen


et al. 1999a). Even if this resistance was identified, isolated and incorporated into commercial cultivars, this would not resolve the problem, because of the existence of polyspecific nematode communities. Changes in the nematode composition will occur, favouring root parasitic nematodes to which these hybrids are not resistant. Also, diffe-rences in the aggressiveness of R. similis have been reported (Sarah et al. 1993, Fallas et al 1995, Marin et al. 1999) when comparing populations. Therefore, broad-spectrum resistance should be sought, conferring resistance to a wide range of known R. similis biotypes, and to the other important parasitic nematodes. Up to now, ‘Yangambi km5’ shows resistance to R. similis (Sarah et al. 1992, Price 1994) and P. coffeae (Viaene et al. 1998), but its progeny produce abnormal leaves and/or erect and semi-erect bun-ches (Stoffelen et al. 1999a). Another source of resistance to these two nematodes is Musa acuminata spp. burmannicoides ‘Calcutta 4’ (Viaene et al. 2000). This variety has already been used in the IITA breeding programme (Swennen and Vuylsteke 1993, Tenkouano et al. 2003), and experiments are on-going.

References Adiko A. 1989 Effect of Meloidogyne incognita on plantain, Musa AAB. International Nematology Network

Newsletter 6:27-30.

Araya M., M. Centeno & W. Carrillo. 1995 Densidades poblacionales y frecuencia de los nematodos parásitos del banano (Musa AAA) en nueve cantones de Costa Rica. Corbana 20:6-11.

Araya M., D. De Waele, & R. Vargas. 2002 Occurrence and population densities of nematode parasites of banana (Musa AAA) roots in Costa Rica. Nematropica 32:21-33.

Araya M., T. Moens & C. Calvo. 2003 Sección de nematología. Pp. 66-111 in Informe anual 2002. Corporación Bananera Nacional, Dirección de Investigaciones.

Barekye A., I.N. Kashaija, E. Adipala & W.K. Tushemereirwe. 1999. Pathogenicity of Radopholus similis and Helicotylenchus multicinctus on bananas in Uganda. Pp. 319-326 in Mobilizing IPM for sustainable banana production in Africa. (E.A. Frison, C.S. Gold, E.B. Karamura and R.A. Sikora, eds). IPGRI-INIBAP, Montpellier, France.

Barekye A., I.N. Kashaija, W.K. Tushemereirwe & E. Adipala. 2000. Comparison of damage levels caused by Radopholus similis and Helicotylenchus multicinctus on bananas in Uganda. Annals of Applied Biology 137:273-278.

Binks R.H. & S.R. Gowen. 1997. Early screening of banana and plantain varieties for resistance to Radopholus similis. International Journal of Nematology 7:57-61.

Boncato A.A. & R.G. Davide. 1992. Radopholus similis on Giant Cavendish banana in Davao del Norte: host range, relative distribution, and density. Pp. 45-52 in Studies on nematodes affecting bananas in the Philippines (R.G. Davide, ed.). Philippine Agriculture and Resources Research Foundation, Inc., Los Baños, Philippines.

Bridge J., R. Fogain & P. Speijer 1997 The root lesion nematodes of banana. Pratylenchus coffeae (Zimmermann, 1898) Filip. & Schu. Stek., 1941. Pratylenchus goodeyi Sher & Allen, 1953. Musa Pest Fact Sheet No. 2. INIBAP, Montpellier, France.

Chau N.N., N.V. Thanh, D. De Waele & E. Geraert. 1997. Plant-parasitic nematodes associated with banana in Vietnam. International Journal of Nematology 7:122-126.

Collingborn F.M.B. & S.R. Gowen. 1997. Screening of banana cultivars for resistance to Radopholus similis and Pratylenchus coffeae. InfoMusa 6(2):3.

Coosemans J., K. Duchateau & R. Swennen. 1994. Root-knot nematode (Meloidogyne spp.) infection on banana (Musa spp.) in vitro. Arch. Phytopath. Pflanz. 29:165-169.

Davide R. 1996. Overview of nematodes as a limiting factor in Musa production. Pp. 27-31 in New Frontiers in Resistance Breeding for Nematode, Fusarium and Sigatoka (E.A. Frison, J.P. Horry and D. De Waele, eds). INIBAP, Montpellier, France.


Davide R.G. 1994. Status of nematode and weevil borer problems affecting banana in the Philippines. Pp. 79-88 in Banana nematodes and weevil borers in Asia and the Pacific. (R.V. Valmayor, R.G. Davide, J.M. Stanton, N.L. Treverrow, and V.N. Roa, eds). INIBAP/ASPNET, Los Baños, La Laguna, Philippines.

Davide R.G. & L.Q. Marasigan. 1985. Yield loss assessment and evaluation of resistance of banana cultivars to the nematode Radopholus similis Thorne and Meloidogyne incognita Chitwood. Philippine Agriculturist 68:335-349.

De Waele D. 2000. Root-knot nematodes. Pp. 307-314 in Diseases of banana, Abacá and Enset (D.R. Jones, ed.). CABI Publishing, Wallingford, UK.

De Waele D. & R.G. Davide. 1998. The root-knot nematodes of banana Meloidogyne incognita (Kofoid & White, 1919) Chitwood, 1949. Meloidogyne javanica (Treub, 1885) Chitwood, 1949. Musa Pest Factsheet No. 2. INIBAP, Montpellier, France.

Fallas G.A., J.L. Sarah & M. Fargette. 1995. Reproductive fitness and pathogenicity of eight Radopholus similis isolates on banana plants (Musa AAA cv. Poyo). Nematropica 25:135-141.

Fallas G.A. & J.L. Sarah. 1995. Effect of temperature on the in vitro multiplication of seven Radopholus similis isolates from different banana producing zones of the world. Fundamental and Applied Nematology 18:445-449.

Fargette M. & P. Quénéhervé. 1988. Populations of nematodes in soil under banana, cv. Poyo, in the Ivory Coast. 1. The nematofauna occurring in the banana producing areas. Revue Nématologie. 11:239-244.

Figueroa A. 1989. Dinamicas poblacionales de cuatro géneros de nematodos parásitos en plátano (Musa AAB, subgrupo plátano cv. Curraré). Asbana 13(32):5-7.

Fogain R., S. Gowen & F. Mekemda. 1996. Screening for susceptibility to Radopholus similis: evaluation of plan-tains AAB and diploid AA, AB, and BB. Tropical Agriculture (Trinidad) 73:281-285.

Fogain R. & S. Gowen. 1995. Pathogenicity on maize and banana among isolates of Radopholus similis from four producing countries of Africa and Asia. Fruits 50:5-9.

Fogain R. 1994. Root knot nematodes: a serious threat to banana productions in Cameroon? MusAfrica 4(3): 3.

Gómez J.T. 1997. Dispersión y niveles de infestación de Radopholus similis Cobb en algunas plantaciones bana-neras del Ecuador. Nematropica 27:111 (abstract).

Gómez J.T. 1980. Determinación de la infestación de fitonematodos en plantaciones bananeras de Urabá, Colombia. Fitopatología Colombiana 9(1):19-32.

Gowen S.R. 2000a. Root-lesion nematodes. Pp. 303-306 in Diseases of Banana, Abacá and Enset (D.R. Jones, ed.). CABI Publishing. Wallingford, UK.

Gowen S.R. 2000b. Spiral nematode. Pp. 306-307 in Diseases of Banana, Abacá and Enset (D.R. Jones, ed.). CABI Publishing. Wallingford, UK.

Gowen S. 1995. Pests. Pp. 382-402 in Bananas and Plantains (S. Gowen, ed.). Chapman & Hall, London, UK.

Gowen S. & P. Quénéhervé. 1990. Nematode parasites of bananas, plantains and abaca. Pp. 431-460 in Plant Parasitic Nematodes in Subtropical and Tropical Agriculture (M. Luc, R.A. Sikora and J. Bridge, eds). CAB International, Wallingford, UK.

Hahn M.L., J.L. Sarah, M. Boisseau, N.J. Vines, D.J. Wright & P.R. Burrows. 1996. Reproductive fitness and pathogenicity of selected Radopholus similis populations on two banana cultivars. Plant Pathology 45:223-231.

Jonathan E.I. & G. Rajendran. 2000a. Pathogenic effect of root-knot nematode Meloidogyne incognita on banana Musa sp. Indian Journal of Nematology 30:13-15.

Jonathan E.I. & G. Rajendran. 2000b. Assessment of avoidable yield loss in banana due to root-knot nematode Meloidogyne incognita. Indian Journal of Nematology 30:162-164.

Jiménez M.F. 1972. Fluctuaciones anuales de la población de Radopholus similis en la zona bananera de Pococí, Costa Rica. Nematropica 2:33-38.

Jiménez F., I. Ma, C. Triviño & L. Velasco. 1998. Influencia de la temperatura y la precipitación en la fluctuación poblacional de Radopholus similis en las zonas Oriental y Central del Ecuador. Pp. 547-555 in Memorias XIII Reunión ACORBAT 1998. (L. Hidalgo Arizaga, ed.). CONABAN, Guayaquil, Ecuador.


Kashaija I.N., R. Fogain & P.R. Speijer. 1999. Habitat management for control of banana nematodes. Pp. 109-118 in Mobilizing IPM for Sustainable Banana Production in Africa. (E.A. Frison, C.S. Gold, E.B. Karamura & R.A. Sikora, eds). INIBAP, Montpellier, France.

Loos C.A. 1962. Studies on the life-history and habits of the burrowing nematode, Radopholus similis, the cause of black-head disease of banana. Proceedings of the Helminthological Society 29:43-52.

López Ch.R. 1980. Determinación de los nematodos fitoparásitos asociados al plátano (Musa acuminata x Musa balbisiana, AAB) en Río Frío. Agronom. Costarr. 4:143-147.

Mani A. & M.S. Al Hinai. 1996. Population dynamics and control of plant parasitic nematodes on banana in the Sultanate of Oman. Nematologia Mediterranae 24:295-299.

Marin D.H., K.R. Barker, D.T. Kaplan, T.B. Sutton & C.H. Opperman. 1999. Aggressiveness and damage potential of Central American and Caribbean populations of Radopholus spp. in banana. Journal of Nematology 31:377-385.

Marin D.H., T.B. Sutton & K.R. Barker. 1998. Dissemination of bananas in Latin America and the Caribbean and its relationship to the occurrence of Radopholus similis. Plant Disease 82:964-974.

McSorley R. 1986. Nematode problems on bananas and plantains in Florida. Nematology Circular No 133. Florida Department Agriculture & Consumer Services, Division of Plant Industry.

McSorley R. & J. L. Parrado. 1986. Helicotylenchus multicinctus on bananas: an international problem. Nematropica 16:73-91.

Minz G., D. Ziv & D. Strich-Harari. 1960. Decline of banana plantations caused by spiral nematodes, in the Jordan Valley, and its control by DBCP. Ktavim 10:147-157.

Moens T., M. Araya, R. Swennen, D. De Waele & J. Sandoval. 2003. Growing medium, inoculum density, exposure time and pot volume: factors affecting the resistance screening for Radopholus similis in banana (Musa spp.). Nematropica 33:9-26.

Moens T., M. Araya & D. De Waele. 2001. Correlations between nematode numbers and damage to banana (Musa AAA) roots under commercial conditions. Nematropica 31:55-65.

Orion D. & M. Bar-Eyal. 1995. Observations on the life cycle and behavior of Helicotylenchus multicinctus in monoxenic culture of Arabidopsis thaliana. Nematropica 25: 67-70.

Orton W.K.J. & M.R. Siddiqi. 1973. Radopholus similis. C.I.H. Description of plant-parasitic nematodes. No. 27. Commonwealth Institute of Helminthology, UK. 4pp.

Patel B.A., R.V. Byas, D.J. Patel & R.S. Patel. 1996. Susceptibility of banana cultivars to root-knot nematodes (Meloidogyne spp.). InfoMusa 5(2):26-27.

Pinochet J., M.C. Jaizme, C. Fernandez, M. Jaumot & D. De Waele. 1998. Screening bananas for root-knot (Meloidogyne spp.) and lesion nematode (Pratylenchus goodeyi) resistance for the Canary Islands. Fundamental and Applied Nematology 21:17-23.

Pinochet J. & P. Rowe. 1979. Progress in breeding for resistance to Radopholus similis on bananas. Nematropica 9:40-43.

Pinochet J. 1978. Histopathology of the root lesion nematode, Pratylenchus coffeae on plantains, Musa AAB. Nematologica 24: 337-340.

Pinochet J. 1977. Occurrence and spatial distribution of root-knot nematodes of bananas and plantains in Honduras. Plant Disease Reporter 61:518-520.

Pone S. 1994. Status of nematode and weevil borer problems affecting banana in some of the Pacific Islands. Pp. 90-105 in Banana Nematodes and Weevil Borers in Asia and the Pacific. (R.V. Valmayor, R.G. Davide, J.M. Stanton, N.L. Treverrow & V.N. Roa, eds). INIBAP/ASPNET, Los Baños, Philippines.

Price N.S. 1994. Field trial evaluation of nematode susceptibility within Musa. Fundamental and Applied Nematology 17:391-396.

Quénéhervé P., P. Cadet, T. Mateille & P. Topart. 1991. Population of nematodes in soils under banana cv. Poyo, in the Ivory Coast. 5. Screening of nematicides and horticultural results. Revue Nématologie 14:231-249.

Razak A.R. 1994. Plant parasitic nematodes, a potential threat to commercial cultivation of banana in Malaysia. Pp. 34-45 in Banana Nematodes and Weevil Borers in Asia and the Pacific. (R.V. Valmayor, R.G. Davide, J.M. Stanton, N.L. Treverrow and V.N. Roa, eds). INIBAP/ASPNET, Los Baños, Philippines.


Reddy P.P. 1994. Status of banana nematodes in India. Pp, 247-254 in Banana Nematodes and Weevil Borers in Asia and the Pacific. (R.V. Valmayor, R.G. Davide, J.M. Stanton, N.L. Treverrow and V.N. Roa, eds). INIBAP/ASPNET, Los Baños, Philippines.

Rodríguez R.R. 1990. Los nematodos de la platanera (Musa acuminata AAA, subgrupo Cavendish Enana) en Canarias (1963-1984). Las Palmas de Gran Canarias, Caja Insular de Ahorros de Canarias. 58pp.

Sánchez R.O. & M. Zuñiga. 2004. Costa Rica, estadísticas de exportación bananera 2003. Corporación Bananera Nacional S.A. (CORBANA). 60pp.

Santor W. & R.G. Davide. 1992. Interrelationship of Radopholus similis and Meloidogyne incognita in banana. Pp. 71-77 in Studies on Nematodes Affecting Bananas in the Philippines (R.G. Davide, ed.). Philippine Agriculture and Resources Research Foundation Inc. Los Baños, Philippines.

Sarah J.L. 2000. Burrowing nematode. Pp. 295-303 in Diseases of Banana, Abacá and Enset (D.R. Jones, ed.). CABI Publishing. Wallingford, UK.

Sarah J.L., J. Pinochet & J. Stanton. 1996. El nematodo barrenador del banano Radopholus similis Cobb. Plagas de Musa – Hoja divulgativa No. 1. INIBAP, Montpellier, France.

Sarah J.L., C. Sabatini & M. Boisseau. 1993. Differences in pathogenicity to banana (Musa sp. cv. Poyo) among isolates of Radopholus similis from different production areas of the world. Nematropica 23:73-79.

Sarah J.L., F. Blavignac, C. Sabatini & M. Boisseau. 1992. Une méthode de laboratoire pour le criblage variétal des bananiers vis-à-vis de la résistance aux nématodes. Fruits 47:559-564.

Sarah J. L. 1989. Banana nematodes and their control in Africa. Nematropica 19:199-216.

Siddiqi M.R. 1972. Pratylenchus coffeae. C.I.H. Description of Plant-Parasitic nematodes. Set 1, No. 6. Commonwealth Institute of Helminthology, UK. 3pp.

Sikora R.A. & E. Schlösser. 1973. Nematodes and fungi associated with root systems of bananas in a state of decline in Lebanon. Plant Disease Reporter 57:615-618.

Speijer P.R. & R. Fogain. 1999. Musa and Ensete nematode pest status in selected African countries. Pp. 99-108 in Mobilizing IPM for Sustainable Banana Production in Africa. (E.A. Frison, C.S. Gold, E.B. Karamura, & R.A. Sikora, eds). INIBAP, Montpellier, France.

Stanton J.M. 1994. Status of nematode and weevil borer problems affecting banana in Australia. Pp. 48-56 in Banana Nematodes and Weevil Borers in Asia and the Pacific. (R.V. Valmayor, R.G. Davide, J.M. Stanton, N.L. Treverrow & V.N. Roa, eds). INIBAP/ASPNET, Los Baños, Philippines.

Stoffelen R., R. Verlinden, J. Pinochet, R.L. Swennen & D. De Waele. 2000. Host plant response of Fusarium wilt resistant Musa genotypes to Radopholus similes and Pratylenchus coffeae. International Journal of Pest Management 46:289-293.

Stoffelen R., R. Verlinden, N.T. Xuyen, R. Swennen & D. De Waele. 1999a. Screening of Papua New Guinea bananas to root-lesion and root-knot nematodes. InfoMusa 8(1):12-15.

Stoffelen R., M.I. Jiménez, C. Dierckxsens, V.T.T. Tam, R. Swennen & D. De Waele. 1999b. Effect of time and inoculum density on the reproductive fitness of Pratylenchus coffeae and Radopholus similis populations on carrot disks. Nematology 1:243-250.

Stoffelen R., V.T.T. Tam, R.L. Swennen & D. De Waele. 1999c. Host plant response of banana (Musa spp.) culti-vars from Southeast Asia to nematodes. International Journal of Nematology 9:130-136.

Swennen R. & D. Vuylsteke. 1993. Breeding black Sigatoka resistant plantains with a wild banana. Tropical Agriculture (Trinidad) 70:74-77.

Tarté R. & J. Pinochet. 1981. Problemas nematológicos del banano, contribuciones recientes a su conocimiento y combate. UPEB, Panamá. 32pp.

Tarté R. 1980. La importancia del conocimiento de la biología y comportamiento de los nematodos parásitos del banano en el desarrollo de métodos eficientes de control. Augura 6:13-21.

Tenkouano A., D. Vuylsteke, J. Okoro, D. Makumbi, R. Swennen & R. Ortiz. 2003. Diploid banana hybrids TMBx5105-1 and TMB2x9128-3 with good combining ability, resistance to black sigatoka and nematodes. HortScience 38:468-472.

Van den Bergh I., D.T.M. Nguyet, N.T. Tuyet, H.H. Nhi & D. De Waele. 2002a. Screening of Vietnamese Musa germplasm for resistance to root knot and root lesion nematodes in the greenhouse. Australasian Plant Pathology 31:363-371.


Van den Bergh I., D.T.M. Nguyet, N.T. Tuyet, H.H. Nhi & D. De Waele. 2002b. Responses of Vietnamese Musa genotypes to Meloidogyne spp. under field conditions. Nematology 4:917-923.

Van den Bergh I., V.T. Tam & N.N. Chau. 2000. Assessment of the occurrence and damage potential of nematodes on bananas in North and Central Vietnam. Pp. 134-149 in Highlights of Musa research and Development in Vietnam. (H.H. Nhi, A. Molina, I. Van den Bergh & P.T. Sen, eds). VASI, Hanoi, Vietnam.

Viaene N., L.F. Durán, J.M. Rivera, J. Dueñas, P. Rowe & D. De Waele. 2003. Responses of banana and plantain cultivars, lines and hybrids to the burrowing nematode Radopholus similis. Nematology 5:85-98.

Viaene N., L.F. Durán, J. Dueñas, J.M. Rivera, D. De Waele & P. Rowe. 2000. Reacción de híbridos y genotipos naturales de Musa al ataque de los nematodos Radopholus similis y Pratylenchus coffeae en ambiente de casa de sombra. P;9 in XIV Reunión de ACORBAT, San Juán, Puerto Rico.

Viaene N., J. Dueñas, J.M. Rivera, P. Rowe & D. De Waele. 1998. Determinación experimental de la reacción a los nematodos Radopholus similis y Pratylenchus coffeae de germoplasma selecto de Musa en casa de sombra. Pp. 43-67 in Informe técnico 1997. Fundación Hondureña de Investigación Agrícola, Programa de Banano y Plátano, La Lima, Honduras.

Wehunt E.J., D.J. Hutchinson & D.I. Edwards. 1978. Reactions of banana cultivars to the burrowing nematode. Journal of Nematology 10:368-370.

The problem of banana root deterioration and its impact on production224 The effect of arbuscular mycorrhizal fungi (AMF)-nematode interactions on root development C.A. Gauggel et al. 225A. Elsen et al.

The effect of arbuscular mycorrhizal fungi (AMF)-nematode interactions on the root development of different Musa genotypesAnnemie Elsen, Rony Swennen and Dirk De Waele1

AbstractIn this study, the effect on Musa plant growth, with focus on the root system, of two arbuscular mycorrhizal fungi (AMF), Glomus mosseae and G. intraracides, and two migratory endoparasitic nematodes, Radopholus similis and Pratylenchus coffeae, were examined. In addition, the AMF-nematode interaction was studied. Ten Musa genotypes with different root systems were selected. Based on their relative mycorrhizal dependency and their host plant response, two genotypes ‘Obino l’ewai’ and ‘Calcutta 4’) were selected for AMF-nematode interaction studies. The experiments were performed under greenhouse conditions.

Mycorrhizal colonization of the roots resulted in significantly better plant growth, even in the presence of nematodes. The effect of AMF on the root system was genotype-dependent and seemed to be related to the relative mycorrhizal dependency of the genotype. Genotypes with a proportionally high secondary and tertiary root weight (like ‘Calcutta 4’ and ‘Kayinja’) had a rather low mycorrhizal dependency and a high fresh root weight. For those genotypes, AMF had no influence on the branching of roots. Genotypes with a proportionally high weight of primary roots (like ‘Obino l’ewai’, ‘Igitsiri’, ‘Mbwazirume’, ‘Pisang lilin’ or ‘Pisang jari buaya’) had a medium to high mycorrhizal dependency and the presence of AMF increased root branching. The nematodes also affected the root system, especially by decreasing branching. Nematode population densities were significantly reduced in the presence of AMF, except for P. coffeae in ‘Obino l’ewai’.

In this study, the effect of the interaction between AMF and nematodes on root morphology was studied for the first time. The analysis is rather complex since many factors are involved. The three factors: genotype, AMF and nematode influence the root system; making firm conclusions very difficult. All together, there is no net effect, since the reduced branching caused by the nematodes, is counter balanced by the increased branching caused by AMF. This could be a possible strategy to reduce the negative impact of nematode infection.

Resumen - Efecto de las interacciones entre hongos micorrícicos arbusculares (HMA) y nematodos en el desarrollo radical de diferentes genotipos de Musa En este estudio, se investigó el efecto de dos hongos micorricicos arbusculares (AMF por sus siglas en inglés). Glomus mosseae y G. Intraradices, y de dos nematodos endoparásitos migratorios, Radopholus similis y Pratylenchus coffeae, sobre el crecimiento de plantas de Musa, con énfasis en su sistema radical. Adicionalmente, se estudió la interacción AMF-nematodo. Se seleccionaron 10 genotipos de Musa con diferentes sistemas radicales. Basado en su dependencia relativa a las micorrizas y su respuesta como planta hospedera, se seleccionaron dos genotipos (‘Calcutta 4’ and ‘Obino l’ewai’) para estudiar la interacción AMF-nematodo. Los experimentos se efectuaron bajo condiciones de invernadero.

La micorrización resultó en un crecimiento significativamente mejor de la planta, aún en presencia de nematodos. El efecto de la AMF sobre el sistema radical dependió del genotipo y pareció estar relacionado con la dependencia relativa del genotipo a las micorrizas. Los genotipos con un peso proporcionalmente alto de raíces secundarias y terciarias (como ‘Calcutta 4’ y ‘Kayinja’) tuvieron

1 Laboratory of Tropical Crop Improvement, Katholieke Universiteit Leuven, Kasteelpark Arenberg 13, 3001 Leuven, Belgium. e-mail: [email protected]


una dependencia bastante baja a las micorrizas y un alto peso de raíces frescas. Para estos genotipos, la AMF no influyó en las ramificaciones. Los genotipos con un peso proporcionalmente alto de raíces primarias (como ‘Obino l’ewai’, ‘Igitsiri’, ‘Mbwazirume’, ‘Pisang lilin’, o ‘Pisang jari buaya’) mostraron una dependencia de mediana a alta a las micorrizas y la presencia de la AMF incrementó las ramificaciones. Los nematodos también afectaron el sistema radical, reduciendo el número de ramificaciones. Las densidades de población de los nematodos se redujeron significativamente en presencia de la AMF, con excepción de P. coffeae en ‘Obino l’ewai’.

En este estudio, se investigó por vez primera, el efecto de la interacción AMF - nematodos en la morfología radical. Este análisis es bastante complejo pues involucra muchos factores. Los tres factores - genotipo, AMF y los nematodos - influenciaron el sistema radical, lo que dificulta establecer conclusiones definitivas. En resumen, no existe un efecto neto, porque la reducción en las ramificaciones causadas por los nemátodos se equilibra con el aumento de las mismas causado por la AMF. Esta podría ser una estrategia para reducir el impacto negativo de la infección por nematodos.

IntroductionRoots function both as a support system and as the nutrient uptake organ of plants. Root morphology changes in response to the soil environment to minimize the metabolic cost of maintaining the root system. In response to nutrient-limiting conditions, plants may increase root fineness, root/shoot ratio or the number or length of the root hairs (Hetrick 1991). Each of these adaptations involves a different metabolic cost. Mycorrhizae are another alternative to such changes.

Several studies have shown that colonization by arbuscular mycorrhizal fungi (AMF) can influence root system morphology of host plants (Tisserant et al. 1992). Most plants infected with AMF develop a denser root system, with a higher number of shorter pri-mary roots of a greater diameter (Berta et al. 1993). In most plants, the branching of roots increased after AMF colonization (Schellenbaum et al. 1991, Tisserant et al. 1992). However, less-branched root systems were observed in mycorrhizal plants of Andropogon gerardii (Hetrick et al. 1988) and Gossypium hirsutum (Price et al. 1989).

Different mechanisms have been suggested for the alteration of root systems following AMF colonization (Berta et al. 1993). In physiological aspects, it is well recognized that AMF can increase nutrient absorption, especially for elements with low mobility in the soil such as phosphorus (P). Other studies have reported changes in phytohormone balance in association with AMF (Dannenberg et al. 1992). Anatomically depressed meristimatic activity of mycorrhizal root apices was found (Berta et al. 1993).

Arbuscular mycorrhizal fungi are able to alter the rooting strategy of plants not only in res-ponse to soil fertility but also in response to other micro-organisms in the rhizosphere (Hetrick 1991). The rhizosphere comprises both beneficial and pathogenic micro-organisms. Migratory endoparasitic nematodes are pathogens infecting roots of many different hosts. Especially in the tropics, they cause huge yield losses (Sasser and Freckman 1987). The infected roots become necrotic due to the nematode, thereby inhibiting nutrient and water uptake and weak-ening the anchorage of the plant. Until now, the most commonly used approach to control these pathogens is the use of nematicides. However, nematicides are expensive and toxic both for the user and the environment. The application of AMF could be a possible approach to nematode management, as a part of an integrated pest management strategy.

In Musa, there are good indications that genotypes that are more resistant to migratory endoparasitic nematodes have a larger and denser root system than the susceptible


genotypes (Blomme 2000, Stoffelen 2000). Until now, only one study reported the effect of AMF on root system morphology of banana plantlets (Jaizme-Vega et al. 1994). The most important modification in root morphology induced in banana plants by AMF consisted of increased branching and increased total root length, which led to a dense rooting pattern. The influence of AMF on the root system of Musa genotypes differing in root morphology and the influence of the AMF/nematode interaction on the root system and vice versa has not been studied.

In this paper, we examine whether AMF can be used as a strategy to manage nemato-des, by inducing a larger and denser root system. The first objective was to investigate whether the effect of AMF on the root system differed for Musa genotypes with ini-tially different root systems. The second objective was to study the interaction between AMF and two plant-parasitic nematodes, R. similis and P. coffeae, on two Musa geno-types with a different root system structure and different responses to AMF (high and low mycorrhizal dependency). The influence of the AMF on the root system and the influence of the altered root system on the nematode reproduction were studied.

Materials and methodsMusa tissue culture plants were obtained from the International Musa germplasm col-lection at the INIBAP Transit Centre (ITC) at K.U.Leuven, Belgium (Table 1). The plant material was proliferated, regenerated and rooted in test tubes on Murashige and Skoog (MS) medium (Murashige and Skoog 1962).

The Glomus mosseae isolate, used in this study, was originally recovered from Pome banana (Musa AAB, Pome sub-group) grown on a biological farm in Los Realejos, Tenerife, Canary Islands, Spain. The Glomus intraradices isolate was recovered from banana in Martinique (MUCL 41833). The isolates were maintained and propagated on

Table 1. Information on Musa genotypes differing in root morphology. Accession name ITC number1 Genome, group

Calcutta 4 0249 AA, burmannicoidesGrande naine 1256 AAA, CavendishIgitsiri 0081 AAA, Mutika/LujugiraKayinja / ABB, Pisang AwakMbwazirume 0084 AAA, Mutika/LujugiraObino lewai 0109 AAB, PlantainPisang lilin 1400 AAPisang jari buaya 0690 AA, Pisang jari buayaYangambi km5 1123 AAA, IbotaGros Michel 1122 AAA, Gros Michel1 ITC number = accession number at the International Musa Germplasm collection at the INIBAP Transit Centre at KULeuven.

sorghum cultures. Two R. similis populations differing in pathogenicity were selected: a highly pathogenic population from Uganda and a less pathogenic population from Indonesia. The P. coffeae population was isolated from banana roots in Ghana. They were maintained in monoxenic conditions on carrot discs at 27+1°C in the dark. Prior to inoculation, the nematodes were isolated using the Baermann funnel technique (Hooper 1990).


In the first three experiments, the micropropagated banana plants were transplanted, after deflasking, to germination trays (60 x 40 cm) for early mycorrhizal colonization during the hardening phase. The substrate consisted of a sterilized 2:1 peat-quartz mixture. Arbuscular mycorrhizal inoculum consisted of 1.5 kg rhizosphere soil from 6-months-old sorghum pot cultures of G. mosseae containing spores, hyphae and hea-vily infected root fragments. The inoculum was spread as a layer between two layers of substrate. The plants not receiving mycorrhizal inoculum received a filtrate of soil inoculum free from AMF propagules. Ten plants were included, per treatment. A period of 6 weeks allowed the establishment of the mycorrhizal symbiosis. After these 6 weeks, eight plants per treatment were transplanted to 1-L containers filled with sterilized 2:1 peat-quartz substrate. For each plant, 1 g Osmocote®, a slow release fertilizer, was added to the substrate. During 16 weeks, i.e. 6 weeks in the germination trays followed by 10 weeks in individual containers, the plants were maintained in the greenhouse at an ambient temperature of 20-27°C, with a 12-hour photoperiod (170-190 µmol m-2s-1 PAR) and a relative humidity of 50-70%.

In the second part, we studied the interaction between AMF and banana nematodes and the selected genotypes were colonized with mycorrhizae as described in the previous paragraph. In the treatments with nematodes, the plants were inoculated with 1000 vermiform (juvenile and adult) nematodes per plant, 8 weeks after planting (2 weeks after transplanting to individual containers). After nematode inoculation, the plants were kept in the greenhouse for 8 weeks (for R. similis) or 10 weeks (for P. coffeae) to allow nematode reproduction.

In the first three experiments, plant growth was assessed 16 weeks after planting. Plant height, number of leaves, foliar surface, fresh and dry shoot weight and fresh root weight were measured for each plant. The relative mycorrhizal dependency (RMD) was determined by expressing the difference between the dry weight of the mycorrhizal plant and the average dry weight of the non-mycorrhizal plant as a percentage of the dry weight of the mycor-rhizal plant (Plenchette et al. 1983). Root growth was characterized by the total root weight, the weight of the primary roots, weight of the secondary and tertiary roots and weight of the in vitro roots. The roots present at time of deflasking (i.e. beginning of the experiment) and thus formed under in vitro conditions are considered as in vitro roots.

The AMF infection was determined in the mycorrhizal plants. For assessing the mycor-rhizal colonization, secondary and tertiary root samples were collected and stained with 0.05% trypan blue in lactic acid (Koske and Gemma 1989). After clarifying, staining and destaining, 20 1-cm fine root segments were mounted on slides and observed under a light microscope. The frequency of AMF colonization (F %) was calculated as the percentage of root segments colonized by either hyphae or arbuscules or vesicles. In the experiments where the interaction between AMF and nematodes was studied, plant growth and mycor-rhizal colonization were assessed as described above. In the treatments inoculated with nematodes, nematode reproduction was assessed (Speijer and De Waele 1997).

The data were analyzed with the STATISTICA® package (Statsoft, 1997). All plant parameters were analyzed by two-way ANOVA, while the mycorrhizal data and nema-tode data were analyzed by one-way ANOVA. Before analysis, data on nematode repro-duction were log(x+1) transformed. Data for frequencies of mycorrhizal colonization were arcsine(x/100) transformed. Means were separated by the Tukey test (P < 0.05).


The relative proportions of the in vitro, primary and secondary and tertiary roots were compared in a Chi-square test (P < 0.05).

ResultsEffect of two AMF on the root development of 10 Musa genotypesTen genotypes were tested in three experiments with ‘Grande naine’ as a reference. Overall, the root colonization with both G. mosseae (experiments 1 + 2) and G. intra-radices (experiment 3) had a beneficial effect on root growth (Table 2). The presence of AMF increased the total fresh root weight by increasing the weight of the first order

Table 2. Influence of Glomus mosseae and G. intraradices on the root growth of Musa genotypes differing in root morphology, 16 weeks after planting. Root fresh weight, g

Total 1st order 2nd and 3rd order In vitro roots

Experiment 1

Without G.mosseae Grande naine 16.9 A 4.7 AB 11.7 A 0.4 ACalcutta 4 18.3 A 5.8 B 12.4 A 0.2 AKayinja 22.0 B 6.2 B 15.4 B 0.5 BObino l’ewai 15.7 A 6.0 A 8.9 A 0.8 ABWith G. mosseae Grande naine 39.3 A 10.7 AB 28.5 A 0.1 ACalcutta 4 36.2 A 10.9 B 25.3 A 0 AKayinja 46.3 B 12.5 B 33.2 B 0.6 BObino lewai 37.2 A 7.7 A 29.5 A 0 AB *** *** *** NS

Experiment 2

Without G. mosseae Grande naine 12.6 a 5.4 bc 6.7 ab 0.53 abcIgitsiri 5.1 a 1.8 a 3.0 a 0.29 aMbwazirume 5.4 a 2.5 ab 2.6 a 0.26 aPisang lilin 10.0 a 4.9 abc 4.8 ab 0.29 aWith G. mosseae Grande naine 45.1 c 13.2 d 31.5 e 0.44 abIgitsiri 29.1 b 5.6 bc 21.5 d 2.01 cMbwazirume 23.5 b 6.2 c 17.1 cd 0.16 aPisang lilin 23.7 b 10.5 d 11.7 bc 1.53 bc

Experiment 3

Without G. intraradices Grande naine 5.75 a 3.16 a 2.59 a /Pisang jari buaya 7.25 a 3.79 a 3.46 a /Yangambi km5 7.50 ab 4.59 a 3.66 a /Gros Michel 6.75 a 3.62 a 3.12 a /With G. intraradices Grande naine 15.75 bc 6.49 b 9.26 bc /Pisang jari buaya 18.88 c 7.44 b 11.44 c /Yangambi km5 15.12 c 6.50 b 8.62 b /Gros Michel 15.87 c 6.31 b 9.56 c /Within a column in an experiment, means followed by the same letter, are not significantly different at P=0.05. Small letters refer to an interaction effect, capital letters refer to a main effect by cultivar. *, **, *** indicate significant dif-ferences at P < 0.05; 0.01; 0.001 for a main effect by AMF. NS indicates no significant differences.


roots and of the secondary and tertiary roots. The weight of in vitro roots (only measu-red in experiment 1 and 2) was not altered by the presence of AMF, except for ‘Igitsiri’ and ‘Pisang lilin’. The number of primary roots tended to increase when AMF were present (data not shown). In general, the AMF had a positive influence on root growth, but the impact differed among genotypes.

After 16 weeks, the proportion of in vitro roots was very small in both mycorrhizal and non-mycorrhizal treatments (Figure 1). The proportion of secondary and tertiary roots was larger than the proportion of primary roots, but some variation among geno-types was observed. The non-mycorrhizal ‘Pisang lilin’ and ‘Mbwazirume’ had the largest proportion of primary roots (49% and 46%, respectively), while the mycorrhizal ‘Igitsiri’ had the smallest proportion of primary roots (19%). The presence of G. mos-seae did not influence the root proportions of ‘Calcutta 4’, ‘Kayinja’ or ‘Pisang lilin’ (Figure 1). For ‘Grande naine’, the results were not very clear. In the first experiment, the AMF did not influence the root proportions, while in the second experiment, the presence of the AMF significantly decreased the proportion of primary root weight. For ‘Obino l’ewai’, ‘Igitsiri’ and ‘Mbwazirume’, the AMF significantly increased bran-ching of the root system as indicated by a proportional decrease of the primary root weight in favour of an increase of the secondary and tertiary root weight.

The presence of G. intraradices significantly increased the proportion of secondary and tertiary roots of all four genotypes tested (Figure 2). In the non-mycorrhizal plants the proportion of primary roots ranged from 52% (‘Pisang jari buaya’) to 61% (‘Yangambi km5’), while in the mycorrhizal plants the proportion of primary roots ranged from 39% (‘Pisang jari buaya’) to 43% (‘Yangambi km5’).

Mycorrhizal colonization with G. mosseae and G. intraradices was successful, as indicated by the high frequencies (Table 3). Differences in relative mycorrhizal depen-

Table 3. Root colonization by Glomus mosseae (exp. 1 and 2) and G. intraradices (exp. 3) and the relative mycorrhizal dependency (RMD) of Musa genotypes differing in root morphology, 16 weeks after planting. Genotype F% I% RMD (%)

Experiment 1

Grande naine 50 a 17 a 66 bcCalcutta 4 61 a 20 a 50 aKayinja 56 a 21 a 58 abObino l’ewai 54 a 15 a 75 c

Experiment 2

Grande naine 72 b 22 b 62 aIgitsiri 70 b 16 ab 83 cMbwazirume 50 a 14 a 70 bPisang lilin 33 a 13 a 70 b

Experiment 3

Grande naine 88 a 28 a 37 aPisang jari buaya 88 a 29 a 60 aYangambi km5 89 a 28 a 48 aGros Michel 93 a 35 a 62 aF% = frequency of the mycorrhizal colonisation in the roots and I% = intensity of the mycorrhizal colonisation in the roots. Within a column in an experiment, means followed by the same letter, are not significantly different at P=0.05.


Prop

ortion

inroot

weigh

tProp

ortion

inroot

weigh

t

-AMF +AMF -AMF +AMF -AMF +AMF -AMF +AMF

-AMF +AMF -AMF +AMF -AMF +AMF -AMF +AMF

10

20

30

40

50

60

70

80

90

100

110

10

20

30

40

50

60

70

80

90

100

110

Calcutta 4 Grande naine Kayinja Obino l'ewai

Grande Naine Igitsiri Mbwazinume Pisang lilin

In vitro roots2nd + 3rd roots1st roots

Figure 1. Effect of Glomus mosseae (AMF) on the proportional root weight of primary (1st), secondary (2nd) and tertiary (3rd) and in vitro roots of four different Musa genotypes 16 weeks after planting (top = experiment 1, bottom = experiment 2)


dency (RMD) were observed in the first and second experiment. ‘Obino l’ewai’ and ‘Calcutta 4’ were selected for further experiments based on their RMD, which were high and low respectively.

Effect of AMF-nematode interactions on root development of two selected Musa genotypesIn experiments 4 and 5, ‘Obino l’ewai’ showed a significantly better plant growth when colonized by G. mosseae (data not shown). The presence of the AMF had a positive influence on root growth, while the presence of R. similis or P. coffeae did not affect root growth in general (Table 4). ‘Obino l’ewai’ colonized with AMF had significantly higher primary, secondary and tertiary root weights. The presence of P. coffeae did not reduce mycorrhizal colonization. The R. similis population density was significantly reduced when AMF were present, while the P. coffeae population density was not affected.

In experiments 6 and 7, ‘Calcutta 4’ showed significantly better plant growth when colonized with G. mosseae (data not shown). The presence of AMF influenced root growth positively, except when plants were infected with R. similis from Indonesia (Table 5). Mycorrhizal ‘Calcutta 4’ plants had significantly higher primary, secondary and tertiary root weights than ‘Obino l’ewai’ (except for plants infected with R. simi-lis from Indonesia). Mycorrhizal colonization was not influenced by the presence of R. similis or P. coffeae, while the population densities of all nematodes tested were significantly suppressed by G. mosseae.

Prop

ortion

inroot

weigh

t

-AMF +AMF -AMF +AMF -AMF +AMF -AMF +AMF20

40

60

80

100Grande naine Pisang jari buaya Yangambi km 5 Gros Michel

2nd + 3rd roots1st roots

Figure 2. Effect of Glomus intraradices (AMF) on the proportional root weight of primary (1st), secondary (2nd) and tertiary (3rd) and in vitro roots of four different Musa genotypes 16 weeks after planting.


For ‘Obino l’ewai’, the presence of R. similis (Ugandan population) significantly redu-ced the branching of the root system in non-mycorrhizal plants, while mycorrhizal colo-nization significantly increased root branching in both control plants and plants infected with R. similis (Ugandan population) (Figure 3). In the non-mycorrhizal ‘Obino l’ewai’

Table 4. Influence of two plant-parasitic nematodes, Radopholus similis and Pratylenchus coffeae, and the AMF, Glomus mosseae, on the root growth of ‘Obino l’ewai’, 16 to 18 weeks after planting. Root fresh weight (g)

Total 1st order 2nd + 3rd In vitro AMF Nematodes

order (F%) per g roots

Experiment 4

-AMF/-rs 9.5 AB 4.2 B 4.7 B 0.6 / /-AMF/+rs Uganda 4.2 A 2.3 A 1.4 A 0.5 / 1336 B-AMF/+rs Indonesia 10.0 B 4.5 B 4.8 B 0.7 / 218 A+AMF/-rs 17.6 AB 6.1 A 10.8 B 0.7 53 b /+AMF/+rs Uganda 16.3 A 6.9 A 7.6 A 1.7 31 a 314 B+AMF/+rs Indonesia 20.1 B 8.7 B 10.4 B 1.3 58 c 74 A *** *** *** ** ***

Experiment 5

-AMF/-pc 4.6 1.2 2.8 B 0.6 / /-AMF/+pc 3.7 1.5 1.5 A 0.7 / 100+AMF/-pc 11.5 3.8 7.5 B 0.2 89 /+AMF/+pc 10.8 3.5 6.6 A 0.7 89 74 ** *** *** NS NS NSWithin columns and experiments, means followed by the same letter are not significantly different at P=0.05. Lower case letters refer to an interaction effect, upper case letters refer to a main effect by nematodes. *, **, *** indicate significant differences at P < 0.05; 0.01; 0.001 for a main effect by AMF. NS indicates no significant differences.

Table 5. Influence of two plant-parasitic nematodes, Radopholus similis (rs) and Pratylenchus coffeae (pc), and the AMF, Glomus mosseae, on the root growth of ‘Calcutta 4’, 16 to 18 weeks after planting. Root fresh weight (g)

Total 1st order 2nd + 3rd In vitro AMF Nematodes

order

Experiment 4

-AMF/-rs 5.4 ab 2.5 ab 2.5 ab 0.4 / /-AMF/+rs Uganda 1.6 a 0.6 a 0.6 a 0.4 / 252 B-AMF/+rs Indonesia 7.6 bc 3.2 b 3.9 ab 0.6 / 171 A+AMF/-rs 13.0 c 4.2 b 8.5 c 0.3 39 /+AMF/+rs Uganda 12.6 b 5.3 b 6.8 bc 0.6 40 142 B+AMF/+rs Indonesia 8.6 bc 4.2 b 3.9 ab 0.5 33 79 A

NS NS *

Experiment 5

-AMF/-pc 2.3 0.6 1.6 0.1 / /-AMF/+pc 2.1 0.8 1.1 0.2 / 234+AMF/-pc 9.5 2.2 6.4 0.9 64 /+AMF/+pc 8.6 2.5 5.3 0.8 48 57 *** *** *** *** NS ***Within columns and experiments, means followed by the same letter are not significantly different at P=0.05. Lower case letters refer to an interaction effect, upper case letters refer to a main effect by nematodes. *, **, *** indicate significant differences at P < 0.05; 0.01; 0.001 for a main effect by AMF. NS indicates no significant differences.


Prop

ortion

inroot

weigh

t

102030405060708090100110

Prop

ortion

inroot

weigh

t

102030405060708090100110

Prop

ortion

inroot

weigh

t

2030405060708090100

110

Prop

ortion

inroot

weigh

t

2030405060708090100110

-AMF/-pc -AMF/+pc +AMF/-pc +AMF/+pc

-AMF/-pc

-AMF/-rs

-AMF/+rs

Uganda

+AMF/-rs

+AMF/+rs

Indonesia

-AMF/+rs

Indonesia

+AMF/+rs

Uganda

-AMF/-rs

-AMF/+rs

Uganda

+AMF/-rs

+AMF/+rs

Indonesia

-AMF/+rs

Indonesia

+AMF/+rs

Uganda

-AMF/+pc +AMF/-pc +AMF/+pc

Calcutta 4

Obino l'ewai

In vitro roots2nd + 3rd roots1st roots

Figure 3. Effect of Glomus mosseae (AMF) and Radopholus similis (rs) or Pratylenchus coffeae (pc) on the proportional root weight of primary(1st), secondary (2nd) and tertiary (3rd) and in vitro roots of ‘Obino l’ewai’ (graphs on the top) and ‘Calcutta 4’ (graphs on the bottom) 16 (when inoculated with R. similis) or 18 weeks (when inoculated with P. coffeae) after planting.


plants, the presence of the P. coffeae caused a significant decrease in the proportion of secondary and tertiary root weight in favour of an increase in the proportion of primary root weight while in the mycorrhizal plants, no shift in proportions of root weight was observed. In absence of nematodes, the proportion of secondary and tertiary root weight remained the same while proportionally the in vitro root weight decreased in favour of the primary root weight. When inoculated with P. coffeae, the AMF significantly increased branching of the root system.

For ‘Calcutta 4’, the presence of R. similis (Ugandan and Indonesian population) did not influence the branching of the root system. Mycorrhizal colonization significantly increased the proportion of secondary and tertiary roots in control plants and plants infected with R. similis (Ugandan population). For R. similis (Indonesian population), no significant effect of the nematode on root morphology was observed. The presence of P. coffeae significantly reduced branching in the non-mycorrhizal treatments but not in the mycorrhizal treatments. The presence of AMF did not alter the proportions of the different root types of ‘Calcutta 4’.

DiscussionEach of the ten Musa genotypes included in our study responded positively to coloniza-tion with G. mosseae or G. intraradices. Our findings are similar to the results obtained by Jaizme-Vega and Azcon (1995), Pinochet et al. (1997) and Yano-Melo et al. (1999) on Musa species with different Glomus species and under different experimental con-ditions. Our study confirms the beneficial effect of colonization with mycorrhizae in the initial phase of plant growth. However, the magnitude of the response varied among genotypes.

A great variation in dependency on mycorrhizal colonization has previously been observed among Musa spp. (AAA group; Declerck et al. 1995). In our study, RMD also differed among genotypes when colonized with G. mosseae: the lowest RMD was observed for ‘Calcutta 4’ and the highest RMD for ‘Obino l’ewai’and ‘Igitsiri’. When colonized with G. intraradices, no differences in RMD were observed. However, for ‘Grande naine’ the RMD appeared to be different depending on the AMF. This illustra-tes that RMD is not only genotype dependent, but also AMF dependent.

The presence of R. similis or P. coffeae did not influence root growth. Previous studies in Musa indicate that R. similis and M. javanica do not influence plant growth at the early stages of infection (Umesh et al. 1988, Pinochet et al. 1997, Elsen et al. 2003). These root pathogens destroy the root system, hampering water and nutrient uptake. Normally, this results in poorer growth of the infected plant, but under these experi-mental conditions, the short time of the experiment was probably not sufficient to detect these negative effects on plant growth.

Glomus mosseae reduced nematode population build-up for both nematode species, regardless of the pathogenicity of the population (with the exception of P. coffeae in ‘Obino l’ewai’). Many reports have demonstrated a decrease in nematode population development resulting in a increased resistance and/or tolerance of R. similis and Pratylenchus spp. (Smith and Kaplan 1988, Vaast et al. 1998). Previous studies in


Musa reported a suppressive effect of AMF on the reproduction of R. similis and P. coffeae (Elsen et al. 2003, Umesh et al. 1988).

Based on our results, it appears that the effect of the AMF on the branching of the root system is genotype-dependent. Genotypes with a proportionally high secon-dary and tertiary root weight (like ‘Calcutta 4’ and ‘Kayinja’) have a rather low mycorrhizal dependency and a high fresh root weight. These findings corroborate the study by Declerck et al. (1995). In addition, the AMF has no influence on the branching of the roots of genotypes with this type of root system. Therefore, it appears that genotypes with a well-developed root system (large, well-branched, good development of root hairs) have no need to establish a mycorrhizal symbiosis. Genotypes with a proportionally high root weight of primary roots (like ‘Obino l’ewai’, ‘Igitsiri’, ‘Mbwazirume’ or ‘Pisang lilin’) have a high to medium mycor-rhizal dependency and the presence of AMF increases branching. In this study, the genotypes representing this type of root system (i.e. high proportion of primary roots), have been cultivated for a long time and are known for their poor anchorage and root growth. It is possible that these genotypes use the AMF symbiosis as a strategy to improve their root growth and the development of their root system.

It is generally assumed that a denser root system has a greater absorbing power than an elongated one. Therefore, not only the network of extra-root mycelium, with its absorbing power and explorative functions, but also the type of root sys-tem could improve the beneficial growth of mycorrhizal plants. Moreover, a much branched root system is particularly useful for Musa, as they are easily uprooted by strong winds, especially when the root system is weakened by the presence of nematodes.

The presence of a highly pathogenic R. similis or a P. coffeae population influenced the root system of both ‘Obino l’ewai’ and ‘Calcutta 4’ as evidenced by reduced branching. Thus the proportion of primary roots increased in mycorrhized plants. However, the total root weight was not affected by the nematodes. The tested R. similis population with a low pathogenicity did not influence root branching. In Musa, Stoffelen (2000) showed that nematodes can infect all root types, including in vitro roots, but they have a preference for the primary roots. In that study, 70 to 90% of the nematodes were extracted from the primary roots, but no reduction of primary root weight was observed. However, the nematode infection caused a significant reduction of the secondary and tertiary root weight (Stoffelen, 2000). Our findings confirm this for P. coffeae infection in ‘Obino l’ewai’ and ‘Calcutta 4’ and for Ugandan R. similis infection in ‘Obino l’ewai’.

In our study, the effect of the interaction between AMF and nematodes on root morphology was studied for the first time. The analysis is rather complex since there are many factors involved. The three factors: genotype, AMF and nematode influence the root system, making conclusions very difficult. All together, there is no net effect, since the reduced branching caused by the nematodes is counterbalan-ced by the increased branching caused by AMF. This could be a possible strategy to reduce the negative impact of nematode infection.


ReferencesBerta G., A. Fusconi & A. Trotta. 1993. VA mycorrhizal infection and the morphology and function of root

systems. Environmental Experimental Botany 33:159-173.

Blomme G. 2000. The inter-dependence of root and shoot development in banana (Musa spp.) under field conditions and the influence of different biotic and abiotic factors on these relationships. Dissertationes de agricultura No. 421, Katholieke Universiteit Leuven, Leuven, Belgium

Dannenberg G., C. Latus, W. Zimmer, B. Hundeshagen, H.J. Schneider-Poetsch & H. Bothe. 1992. Influence of vesicular-arbuscular mycorrhiza on phytohormone balances in maize (Zea mays L.). Journal of Plant Physiology 141:33-39.

Declerck S., C. Plenchette & D.G. Strullu. 1995. Mycorrhizal dependency of banana (Musa acuminata, AAA group) cultivar. Plant and Soil 176:183-187.

Elsen A., H. Baimey, R. Swennen & D. De Waele. 2003. Relative mycorrhizal dependency and mycorrhiza-nematode interaction in banana cultivars (Musa spp.) differing in nematode susceptibility. Plant and Soil 256:303-313.

Hetrick B.A.D. 1991. Mycorrhizas and root architecture. Experientia 47:355-362.

Hetrick B.A.D., J.F. Leslie, G.T. Wilson & D.G. Kitt. 1988. Physical and topological assessment of effects of vesicular-arbuscular mycorrhizal fungus on root architecture of big bluestem. New Phytologist 110:85-96.

Hooper D.J. 1990. Extraction and processing of plant and soil nematodes. Pp. 45-68 in Plant Parasitic Nematodes in Subtropical and Tropical Agriculture (L. Luc, R.A. Sikora, & J. Bridge, eds). CAB International, Wallingford, UK.

Jaizme-Vega M.C. & R. Azcon. 1995. Response of some tropical and subtropical cultures to endomycorrhizal fungi. Mycorrhiza 5:213-217.

Jaizme-Vega M.C., G. Berta & S. Gianinazzi. 1994. Effect of Glomus intraradices on root system morphology of micropropagated banana plants. In Abstracts of the fourth symposium on mycorrhizas, 11-14 July 1994, Granada, Spain.

Koske R.E. & J.N. Gemma. 1989. A modified procedure for staining roots to detect VA mycorrhizas. Mycological Research 92:486-488.

Murashige T. & F. Skoog. 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiologia Plantarum 15:473-497.

Pinochet J., C. Fernandez, M.C. Jaizme-Vega & P. Tenoury. 1997. Micropropagated banana infected with Meloidogyne responds to Glomus intraradices and phosphorus. HortScience 32:101-103.

Plenchette C., J.A. Fortin & V. Furlan. 1983. Growth response of several plant species to mycorrhizae in a soil of moderate P-fertility. I. Mycorrhizal dependency under field conditions. Plant and Soil 70:199-209.

Price N.S., R.W. Roncadori & R.S. Hussey. 1989. Cotton growth as influenced by phosphorus nutrition and vesicular-arbuscular mycorrhizas. New Phytologist 111:61-66.

Sasser J.N. & D.W. Freckman. 1987. A world perspective on nematology: the role of the society. Pp. 7-14 in Vistas on nematology (J.A. Veech & D.W. Dickson, eds). Society of Nematologists, Hyatsville, USA.

Schellenbaum L., G. Berta, F. Ravolanira, B. Tisserant., S. Gianinazzi & A.H. Fitter. 1991. Influence of endomy-corrhizal infection on root morphology in micro-propagated woody plant species (Vitis vinifera L.). Annals of Botany 68:135-141.

Smith G.S. & D.T. Kaplan. 1988. Influence of mycorrhizal fungus, phosphorus and burrowing nematode inte-ractions on growth of rough lemon citrus seedlings. Journal of Nematology 20:539-544.

Speijer P.R. & D. De Waele. 1997. Screening of Musa germplasm for resistance and tolerance to nematodes. INIBAP Technical Guidelines No. 1. INIBAP, Montpellier, France.

Statsoft Inc. 1997. STATISTICA release 5. Statsoft, Tulsa, USA.

Stoffelen R. 2000. Early screening of Eumusa and Australimusa bananas against root-lesion and root-knot nema-todes. Dissertationes de agricultura No. 426. Katholieke Universiteit Leuven, Leuven, Belgium.

Tisserant B., L. Schellenbaum, V. Gianinazzi-Pearson & S. Gianinazzi. 1992. Influence of infection by an endo-mycorrhizal fungus on root development and architecture in Platanus acerfolia. Allonia 30:1171-1181.


Umesh K.C., K. Krishnappa & D.J. Bagyaraj. 1988. Interaction of burrowing nematode, Radopholus similis (Cobb, 1893) Thorne 1949, and VA mycorrhiza, Glomus fasciculatum (Thaxt.) Gerd. and Trappe in banana (Musa acuminata colla.). Indian Journal of Nematology 18:6-11.

Vaast P., E.P. Cashwell-Chen & R.J. Zasoski. 1998. Influences of a root-lesion nematode, Pratylenchus coffeae and two arbuscular mycorrhizal fungi, Acaulaspora mellea and Glomus clarum on coffee (Coffea arabica L.). Biology and Fertility of Soils 26:130-135.

Yano-Melo A.M., O.J. Saggin, N.F. Melo & L.C. Maia. 1999. Effects of arbuscular mycorrhizal fungi on the acclimatization of micropropagated banana plantlets. Mycorrhiza 9:119-123.

The problem of banana root deterioration and its impact on production238 Secondary metabolites in roots and implications for nematode resistance in banana C.A. Gauggel et al. 239N. Wuyts et al.

Secondary metabolites in roots and implications for nema-tode resistance in banana (Musa spp.)Nathalie Wuyts1*, Georges Lognay2, László Sági1, Dirk De Waele1 and Rony Swennen1

AbstractSecondary metabolites and more specifically phenylpropanoids, in roots of Musa have been related to nematode resistance based on the differential presence of these compounds in the roots of resistant and susceptible cultivars. However, a strong biochemical basis for resistance has not yet been elucidated. The current study demonstrated biochemical differences in the phenylpropanoid pathway of secondary metabolism between cultivars resistant to the banana nematode Radopholus similis, ‘Yangambi km5’ and ‘Pisang jari buaya’, and the susceptible ‘Grande naine’. Unlike previously reported results, lignification and preformed phenolic cells were not found to be specific features of the resistant cultivars. Accumulation of sinapic acid in the thickened cell walls of the endodermis and vascular elements was induced by nematode infection in both the resistant and the susceptible cultivars, while flavonoid-containing cells in the vascular cylinder were only found in resistant cultivars after infection. ‘Grande naine’ roots contained the highest total phenolic content, which could be related to high amounts of proanthocyanidins. In ‘Yangambi km5’ roots, the proanthocyanidin and total phenolic content was the lowest, but mean amounts of total phenols increased after infection, unlike in the other cultivars.

Resumen - Metabolitos secundarios en raíces y sus implicaciones para la resistencia a nematodos en banano (Musa spp.)Metabolitos secundarios y más específicamente los fenilpropanoides, en sistemas radicales de Musa, han sido relacionados a la resistencia a nematodos, basados en la presencia diferencial de estos compuestos en las raíces de cultivares resistentes y susceptibles. Sin embargo, todavía no se cuenta con una fuerte base bioquímica que explique esta resistencia. El presente estudio muestra diferencias bioquímicas en la vía del metabolismo secundario del fenilpropanoide entre cultivares resistentes al nematodo barrenador del banano Radopholus similis, ‘Yangambi km5’ y ‘Pisang jari buaya’, y el susceptible ‘Grande naine’. A diferencia de reportes anteriores, no se encontró que la lignificación y las células fenólicas preformadas estén específicamente relacionadas con los cultivares resistentes. La acumulación de ácido sinapico en el engrosamiento de las paredes celulares de la endodermis y elementos vasculares, fue inducido por la infección de nematodos tanto en los cultivares resistentes como susceptibles, mientras que células conteniendo flavonoides en el cilindro vascular fueron encontradas únicamente en cultivares resistentes después de la infección. Las raíces de ‘Grande naine’ mostraron el mayor contenido de fenoles, lo que pudo haber estado relacionado con altos niveles de proantocianidinas. En las raíces de ‘Yangambi km5’, la proantocianidina y el contenido total fenólico fue el mas bajo, pero los niveles totales promedio fenólicos aumentaron después de la inoculación, a diferencia de los otros cultivares.

IntroductionIn the past fifteen years, various sources of resistance and tolerance to parasitic nema-todes have been identified within banana. However, the underlying mechanisms, which make it possible for these plants to suppress nematode reproduction or suffer little

1 Laboratory of Tropical Crop Improvement, Catholic University of Leuven (KULeuven), Kasteelpark Arenberg 13, 3001 Leuven, Belgium, 2 Unit of General and Organic Chemistry, Gembloux Agricultural University, Passage des déportés 2, 5030 Gembloux, Belgium.*Corresponding author, e-mail: [email protected]


injury are, as in many other plant species, poorly understood. Knowledge of these mechanisms is nevertheless of critical importance to develop strategies for the control of nematodes, especially when one wants these strategies to be environmentally-safe and suitable for subsistence banana growers. Such insights can provide ‘resistance markers’ to facilitate screening of Musa germplasm and provide Musa breeders and molecular biologists with the necessary principles to create nematode resistant plants.

Plants interact with their environment through secondary metabolism. Within this part of the metabolism, the phenylpropanoid pathway is of critical importance as its products (phenolic compounds) protect the plant against abiotic and biotic factors (reviewed by Dixon and Paiva 1995). To date, resistance against nematodes in Musa has been correlated with phenylpropa-noids, i.e. lignification in ‘Pisang jari buaya’ (AA, ‘Pisang jari buaya’ group) (Fogain and Gowen 1996), preformed phenolic cells and infection-induced accumulation of phenolics in ‘Yangambi km5’ (AAA, Ibota group) (Fogain and Gowen 1996, Valette et al. 1998), phenyl-phenalenone phytoalexins in Musa acuminata (Binks et al. 1997, Luis 1998) and high levels of proanthocyanidins in ‘Kunnan’ (AB) (Collingborn et al. 2000). However, these metabolites were not characterized to a sufficient biochemical level and/or no reference was made to their physiological relevance in nematode resistance. Research on secondary metabolites in banana roots has been very limited and provides no guidance for crop improvement.

The results presented here deal with the biochemical nature of nematode resistance encountered in ‘Pisang jari buaya’ and ‘Yangambi km5’, in comparison with the suscep-tible ‘Grande naine’ (AAA, Cavendish group). Various techniques were used, including histochemical staining of root sections and spectrophotometric detection of phenols in root extracts. Results obtained so far are not conclusive about the biochemical basis for resis-tance, but provide guidance for future research into the secondary metabolism of Musa.

Materials and methodsPlant material, nematodes and experimental set-upPlant material was obtained from the International Musa germplasm collection at the INIBAP Transit Centre at KULeuven. It included ‘Grande naine’ (ITC1256) (GN), which was chosen as a susceptible cultivar (Stoffelen et al. 2000), and ‘Yangambi km5’ (ITC1123) (YKm5) and ‘Pisang jari buaya’ (ITC0312) (PJB), which were chosen as resistant cultivars. Micropropagated banana plants were planted in pots filled with 600 ml of a sterilized 2:1 peat-quartz mixture. To each plant 1 g Osmocote®, a slow release fertilizer, was added.

A highly virulent strain of R. similis from banana plants in Uganda was used in the experiment (Fallas et al. 1995). After ten weeks of growth in the greenhouse, eight plants of each cultivar were inoculated with 1000 vermiform (female and juvenile) nematodes per plant. Eight uninfected plants were included per cultivar. Histochemical staining of root sections and analysis of root extracts was performed two weeks after infection.

Histochemical staining of root sectionsRoot sections were hand cut from fresh root samples. For the detection of phenylpropa-noids, sections were stained within 2 min with saturated (0.25%, w/v) diphenylboric


acid-2-aminoethyl ester (DPBA) (Sigma-Aldrich Inc., Bornem, Belgium) in MilliQ water containing 0.02% (v/v) Triton-X-100. The sections were visualized imme-diately with an epifluorescence microscope equipped with a DAPI filter (excitation 340-380 nm, suppression LP 430 nm) and a FITC filter (excitation 450-490 nm, suppression LP 520 nm). Identification was made by comparing colour and intensity to standard references (Peer et al. 2001).

For the histochemical staining of lignin in root sections, hand cut sections were fixed in 4% glutaraldehyde for 60 min and rinsed with water. For Mäule staining, fixed sec-tions were immersed in 0.5% KMnO4 for 10 min, rinsed with water and destained with 10% HCl for 5 min. After a final rinse they were mounted in concentrated NH4OH and examined by bright-field microscopy. For Wiesner staining, fixed sections were stained with 10% phloroglucinol in ethanol/water (95/5, v/v) for 2 min, mounted in concentrated HCl (37%) and examined by bright-field microscopy (Strivastava 1966). No specific staining technique was used for the detection of suberin. However, as sube-rin has a lignin-like structure in its polyaromatic domain, it fluoresces and is stained by the Wiesner and Mäule reagents. Photographic documentation of root sections was achieved with a SPOT RT CCD camera and SPOT software version 3.3 (Diagnostic Instruments Inc., USA).

Extraction and analysis of phenols For the extraction of phenols from roots, approximately 1-2 g of freshly harvested roots was ground in liquid nitrogen. Proanthocyanidin (condensed tannins) amounts were determined using the butanol/HCl assay described by Collingborn et al. (2000) for banana root samples. Ground samples (100 mg) were treated with 5 ml 1-butanol/HCl (95/5, v/v), mixed and incubated at 95°C. After 1 h, samples were thoroughly mixed and centrifuged at 3000 g for 10 min. The absorbance of the top phase was read at 550 nm against a blank sample using a Novaspec II spectrophotometer (Amersham Pharmacia Biotech Inc., Buckinghamshire, UK).

The total phenolic content of root samples was determined using the Folin-Ciocalteu assay (Singleton and Rossi 1965). Ground samples (1 g) were extracted by shaking continuously in a 2 ml aqueous methanol (50%) solution for 1 h at room temperature. Extracts were filtered through a 0.45 µm pore PTFE syringe filter (Merck, Darmstadt, Germany) and stored at -20°C until analysis. In glass tests tubes, 200 µl of the root extracts was mixed with 5 ml water. To each of the samples 500 µl of the Folin-Ciocalteu reagent (Merck, Darmstadt, Germany) was added. Samples were thoroughly mixed and after 3 min, 1 ml saturated sodium carbonate solution (35% in water, w/v) was added. Samples were further diluted to 10 ml with water and left in the dark at room temperature for 1 h. Absorbance was measured at 727 nm against water using a Novaspec II spectrophotometer. A standard calibration curve of gallic acid was determined and results were expressed as gallic acid equivalents/g fresh roots.

Statistical analysisData on the phenolic content of roots were analysed with the STATISTICA® package (Anonymous 2001). Since the number of samples was small (five per cultivar for each treatment) and variances not homogeneous (Levene’s test for homogeneity of varian-


ces), the non-parametric equivalent of ANOVA, Kruskal-Wallis analysis of variance by ranks, was applied. If Kruskal-Wallis was significant, the significance of the differences between treatments and cultivars was pair wise calculated as described by Siegel and Castellan (1988).

Results and discussionHistochemical staining of root sectionsLignified root cell wallsThe DPBA-staining technique was used for the localization and identification of phenolic compounds in root sections. Syringyl lignin units (derived from sinapic acid) and guaiacyl lignin units (derived from caffeic acid) appeared blue and greenish-white respectively when viewed through a DAPI filter (Figure. 1A, F) and green and bright yellow respectively through a FITC filter (Figure 1B, C, D).

Various studies on resistance mechanisms in banana roots deal with the hypothesis that PJB constitutively contains higher amounts of lignin (Pinochet 1988, Binks 1996, Fogain and Gowen 1996, Elsen et al. in press). Among the cultivars tested earlier, PJB and ‘Calcutta 4’ had the highest lignification of the central cylinder, while in the sus-ceptible cultivars (including GN) and YKm5, no or very few cell walls were lignified (Fogain and Gowen 1996, Elsen et al. in press). In the present experiment however, the level of lignification did not differ between the cultivars. In roots of the susceptible GN, cells of the endodermis and central cylinder showed the same degree of thickening of their walls as those of the resistant PJB and YKm5 (Figure 1A-D). In individual roots of the three cultivars, lignification of cell walls increased with the age of the root and the development of laterals. Roots that had recently emerged from the corm were not ligni-fied (Figure 1E, F), while older roots showed the strongest cell wall thickening (Figure 1A-D). This observation might explain the contradictory reports on the occurrence of nematodes in the central cylinder of banana roots (Valette et al. 1997 and 1998, Elsen et al. in press). The lignified anticlinal walls of the endodermis prevent nematodes from penetrating the central cylinder and affecting the vascular elements (Blake 1961). But in young roots, nematodes can be found in the vascular tissue (Mateille 1994, Sarah et al. 1996) because the barrier – the lignified walls of the endodermis – is not developed yet.

In YKm5, the thickened endodermis was predominantly composed of guaiacyl units (yellow fluorescence) (Figure 1D), while in PJB and GN syringyl units (green fluores-cence) prevailed (Figure 1A-C). This observation was confirmed by staining with the Wiesner and Mäule reagents which are specific for total lignin (guaiacyl and syringyl units) and syringyl units respectively. In the three cultivars, guaiacyl units were the dominant constituents of lignified cell walls inside the central cylinder. The suberized walls of the outer cortical cells contained predominantly sinapic acid in the case of PJB and GN and caffeic acid in the case of YKm5.

In response to nematode infection, syringyl lignin concentrations increased in the endodermis of YKm5 and in the central cylinder of GN, YKm5 and PJB. This reac-tion seems to be part of the general defence response of banana, as it occurred in both resistant and susceptible cultivars and in response to both nematodes and fungi


Figure 1. A-D - Cross sections of an older root of ‘Grande naine’ (A, B), ‘Pisang jari buaya’ (PJB) (C) and ‘Yangambi km5’ (D) showing the strong lignification of the endodermis and central cylinder (100x). E, F - Cross sections of young PJB roots showing the absence of, or weak lignification of, the endodermis and cen-tral cylinder (100x). G - Flavonol-containing cells (arrows) in the cortex of a PJB root (100x). H - Cross sec-tion through a nema-tode-damaged PJB root showing the presence of flavonols (arrows) in the tissue borde-ring the necrotic area (100x). Sections were stained with DPBA and viewed under a DAPI filter (A, F) or a FITC filter (B-E, G, H). En: endodermis, PP: protophloem, MP: met-aphloem, PX: protoxy-lem, MX: metaxylem.


(De Ascencao and Dubery 2000). However, the rate of increase and the total amount could be the decisive factor for resistance/tolerance, as was observed in Fusarium wilt-tolerant Musa (De Ascencao and Dubery 2000). Syringyl lignin could be more resistant to enzymatic degradation by pathogens; or compounds released after wall degradation could be toxic to the invaders. In in vitro bioas-says, sinapic acid had a low toxicity and did not repel or inhibit egg hatch of R. similis (Wuyts, unpubl. data).

Phenol-containing root cells Phenolic compounds were detected in cells as granules or as an amorphous mass (Figure 1G), as was described by Mueller and Beckman (1974). They were identified as the flavonols quercetin and kaempferol by their characteristic fluorescence, which was golden yellow and bright green when viewed with a FITC-filter. The lowest num-ber of phenolic cells was found in the cortex of YKm5, while PJB contained the most. This is not in agreement with the proposed mechanism of constitutive resistance in YKm5 (Fogain and Gowen 1996). In the cortex of infected roots, the number of fla-vonol-containing cells did not increase in the three cultivars, which corresponds with the findings of Elsen et al. (in press) for YKm5 in in vitro screenings for resistance. In the central cylinder of YKm5 and PJB, however, flavonoid-containing cells were pre-sent when necrosis had developed in the cortex. Preliminary results of HPLC-analysis of root extracts, confirmed that infected roots of YKm5 and PJB contained higher concentrations of quercetin and kaempferol, while in GN a decrease was found. Flavonols could function in the central cylinder as protective agents or as regulators of physiological processes related to infection, or defence against infection.

Cells bordering the nematode-damaged tissue in the cortex contained flavonols in their walls and in the extracellular spaces in GN, PJB and YKm5 (Figure 1H). The accumu-lation of phenolic compounds, including flavonols, at the site of infection or wounding is a general feature of the plant response to biotic factors (Dixon and Paiva 1995). In in vitro bioassays quercetin and kaempferol were not toxic to R. similis, but they did repel the nematode at concentrations as low as 50 ppm (Wuyts N., unpubl. data). Flavonols may also act as phytoalexins against secondary infection by pathogenic soil organisms which use the damaged tissue as an ‘easy’ access to the root.

Analysis of phenol content in root extracts Proanthocyanidins were analysed in roots of susceptible and resistant cultivars to determine their role in the response to nematode infection (Table 1). Amounts were not significantly different between cultivars, but the highest values were obtained for GN and the lowest for YKm5. Nematode infection had no significant effect on proantho-cyanidin accumulation, although the increase was the highest in PJB. Total phenolic content of roots did not differ significantly between cultivars or between treatments. The high concentration in GN roots could be related to the high concentrations of anthocyanidin-related compounds. Only in YKm5 roots (with the lowest constitutive amounts of total phenolics) did the mean phenolic concentration increase in nematode infected roots, which cannot be related to the proanthocyanidins. HPLC-analysis of root extracts is necessary to determine the absolute amounts of the different phenolic compounds in roots.


Collingborn et al. (2000) have suggested the butanol/HCl-assay as a rapid test in screening for resistance against R. similis, based on their observation that the resistant ‘Kunnan’ contained significantly higher amounts of proanthocyani-dins pre- and post-infection compared with the susceptible ‘Dwarf Cavendish’. However, in the current experiment as well as in the findings of Binks (1996), no correlation existed between resistance and proanthocyanidin amounts. Moreover, the amounts were not significantly higher in roots of infected plants compared with uninfected plants in either the resistant or the susceptible cultivar. However, in the experiment, a random sample of roots was taken from infected and unin-fected plants, while Collingborn et al. (2000) specifically selected areas with lesions for extraction. In general, proanthocyanidins accumulate in response to wounding by the action of polyphenoloxidases (browning reaction). As a result, higher amounts of proanthocyanidins can be expected in roots with lesions than in undamaged ones. Furthermore, in susceptible cultivars the relative accumulation of proanthocyanidins is higher after nematode infection than in resistant ones (data of Collingborn et al. 2000). Proanthocyanidins have strong negative effects on herbivorous insects through their protein-binding capacity, which results in reduced efficiency of nutrient absorption in gut and midgut lesions (Hagerman and Butler 1991). To determine whether proanthocyanidins can be considered as phytoalexins against nematodes in banana roots, they need to be tested in bio-assays with nematodes.

ConclusionsThe aim of the study was to elucidate the biochemical basis for nematode resistance. Constitutive lignification of root cell walls, and induced cell wall strengthening upon nematode infection, were similar in susceptible and resistant cultivars. In resistant cul-tivars, the number of cells containing flavonols in the central cylinder was higher when the cortex was damaged by nematodes. The absolute amount of proanthocyanidins in roots cannot be considered as a ‘biochemical’ marker for nematode resistance, as the highest amounts were found in the susceptible cultivar. Future research includes the analysis of root extracts by HPLC and a search for the physiological relevance of fla-vonoids in nematode-infected roots.

Table 1. Amount of proanthocyanidins and total phenolic compounds determined by spectrophotometry in roots of three banana cultivars either uninfected or infected by R. similis (two weeks after inoculation). Proanthocyanidinsa Total phenolic contenta

(A550 nm) (µg GAEb/g FW)

Cultivar - R. similis + R. similis - R. similis + R. similis

Grande naine 0.471 0.486 (+ 3%) 230 224 (- 3%)Yangambi km5 0.279 0.311 (+ 11%) 157 205 (+ 31%)Pisang jari buayac 0.335 0.446 (+ 33%) 246 240 (- 2%)a results were not significantly different (P < 0.05) between cultivars and treatments (GN and YKm5) (n = 5) as calculated by the Kruskal-Wallis analysis of variance by ranks; b gallic acid equivalents; c n = 1, not included in the statistical analysis.


AcknowledgementsThis work was supported by a grant from the ‘Institute for the Promotion of Innovation through Science and Technology in Flanders’ (IWT-Vlaanderen), Belgium.

ReferencesAnonymous. 2001. STATISTICA® Release 6. Statsoft Inc., Tulsa, USA.

Binks R.H. 1996. Aspects of biochemical resistance in Musa spp. Nematropica 26:243.

Binks R.H., J.R. Greenham, J.G. Luis & S.R. Gowen. 1997. A phytoalexin from roots of Musa acuminata var. ‘Pisang sipulu’. Phytochemistry 45:47-49.

Blake C.D. 1961. Root rot of bananas caused by Radopholus similis (Cobb) and its control in New South Wales. Nematologica 6:295-310.

Collingborn F.M.B., S.R. Gowen & I. Mueller-Harvey. 2000. Investigations into the biochemical basis for nematode resistance in roots of three Musa cultivars in response to Radopholus similis infection. Journal of Agricultural and Food Chemistry 48:5297-5301.

De Ascensao A.R.D.C.F. & I.A. Dubery. 2000. Panama disease: cell wall reinforcement in banana roots in response to elicitors from Fusarium oxysporum f.sp. cubense race four. Phytopathology 90:1173-1180.

Dixon R.A. & N.L. Paiva. 1995. Stress-induced phenylpropanoid metabolism. Plant Cell 7:1085-1097.

Elsen A., J. Orajay & D. De Waele. 2004. Expression of nematode resistance in in vitro roots of three Musa genotypes in response to Radopholus similis. in Proceedings of the Fourth International Congress of Nematology held in Tenerife, Spain, 8-13 June 2002. Nematology. Monographs and Perspectives 2 (R.C. Cook & D.J. Hunt, eds). Brill Academic Publishers, Leiden, The Netherlands (in press).

Fallas G.A., J.L. Sarah & M. Fargette. 1995. Reproductive fitness and pathogenicity of eight Radopholus similis isolates on banana plants (Musa AAA cv. Poyo). Nematropica 25:135-141.

Fogain R. & S.R. Gowen. 1996. Investigations on possible mechanisms of resistance to nematodes in Musa. Euphytica 92:375-381.

Hagerman A.E. & L.G. Butler 1991. Tannins and lignins. Pp. 355-387 in Herbivores: Their Interaction with Secondary Plant Metabolites, Vol. I (G.A. Rosenthal & M.R. Berenbaum, eds). Academic Press, San Diego, USA.

Luis J.G. 1998. Phenylphenalenone-type phytoalexins and phytoanticipins from susceptible and resistant cultivars of Musa species. Its potential for engineering resistance to fungi and nematodes into banana. Acta Horticulturae 490:425-430.

Mateille T. 1994. Biochemical reactions in roots of Musa acuminata (AAA Cavendish group) cv ‘Poyo’ and ‘Gros Michel’ infested by three nematodes. Fundamental and Applied Nematology 17:283-290.

Mueller W.C. & C.H. Beckman. 1974. Ultrastructure of the phenol-storing cells in the roots of banana. Physiological Plant Pathology 4:187-190.

Peer W.A., D.E. Brown, B.W. Tague, G.K. Muday, L. Taiz & A.S. Murphy. 2001. Flavonoid accumulation patterns of transparent testa mutants of Arabidopsis thaliana. Plant Physiology 126:536-548.

Pinochet J. 1988. A method for screening banana and plantains for lesion forming nematodes. Pp. 62-65 in Nematodes and the borer weevil in bananas: present status of research and outlook. Proceedings of a workshop held in Bujumbura, Burundi, 7-11 December 1987. INIBAP, Montpellier, France.

Sarah J.L., J. Pinochet & J. Stanton. 1996. The burrowing nematode of bananas, Radopholus similis Cobb, 1913. Musa Pest Fact Sheet No. 1. INIBAP, Montpellier, France.

Siegel S. & N.J. Castellan. 1988. Nonparametric statistics for the behavioural sciences. McGraw-Hill, Tuas Basin Link, Singapore.

Singleton V.L. & J.A. Rossi. 1965. Colorimetry of total phenolics with phosphomolybdic-phosphotungstic acid reagent. American Journal of Enology and Viticulture 16:144-158.

Stoffelen R., R. Verlinden, J. Pinochet, R.L. Swennen & D. De Waele. 2000. Host plant response of Fusarium wilt resist-ant Musa genotypes to Radopholus similis and Pratylenchus coffeae. International Journal of Pest Management 46:289-293.

The problem of banana root deterioration and its impact on production246 Secondary metabolites in roots and implications for nematode resistance in banana C.A. Gauggel et al. 247

Strivastava L.M. 1966. Histochemical studies on lignin. Tappi 49:173-183.

Valette C., M. Nicole, J.L. Sarah, M. Boisseau, B. Boher, M. Fargette & J.P. Geiger. 1997. Ultrastructure and cytochem-istry of interactions between banana and the nematode Radopholus similis. Fundamental and Applied Nematology 21:65-77.

Valette C., C. Andary, J.P. Geiger, J.L. Sarah & M. Nicole. 1998. Histochemical and cytochemical investigations of phe-nols in roots of banana infected by the burrowing nematode Radopholus similis. Phytopathology 88:1141-1148.

The problem of banana root deterioration and its impact on production246 Secondary metabolites in roots and implications for nematode resistance in banana C.A. Gauggel et al. 247

6

Recommendations

Recomendaciones

C.A. Gauggel et al. 249Recommendations - Recomendaciones

RecommendationsFuture efforts in research and development should address the following issues:

• Biological, physical and chemical causes of soil fatigue and their long term implications.

• Key indicators to determine soil quality and health.

• Study the interactions in the rhizosphere among microorganisms and soil elements.

• Study the practical value of mycorrhizae, endophytes and other microorganisms in the field.

• Determine the potential use of organic amendments.

• Determine the strategies with collaborators for the development of these ideas.

RecomendacionesLas iniciativas futuras en investigación y desarrollo deben enfocar los temas siguientes:

• Factores biológicos, físicos y químicos que ocasionan la fatiga del suelo y sus implicaciones a largo plazo.

• Principales indicadores para determinar la calidad y salud del suelo.

• Estudio de las interacciones a nivel de rizosfera entre los microorganismos y los elementos del suelo.

• Investigar el valor práctico de las micorrizas, endofitos y otros microorganismos en el campo.

• Determinar el uso potencial de las enmiendas orgánicas.

• Definir las estrategias con los colaboradores para el desarrollo de estas ideas.


7

List of participants

Lista de participantes

C.A. Gauggel et al. 253List of participants - Lista de participantes

Argentina

Ranea DiegoEstrada 845 Acassuso, Buenos AiresTel.: 5411 4733 0039Fax: 5411 4733 [email protected]

Australia

Pattison TonyDept. Primary Ind., Center for Wet Tropics Ag. P.O. Box 20, South Johnstone Qld. 5849Tel.: 61 7 4064 1127Fax: 61 7 4064 [email protected]

Turner DavidSchool of Plant Biology M084,Faculty of Nat. and Agric. Scien., University of Western Australia, 35 Stirling Hwy, Crawley WA 6009, PerthTel.: 61 8 9380 2415Fax: 61 8 9380 [email protected]

Belgium

Delvaux Bruno Unité ECOP-GC Université Catholique de Louvain Croix du Sud 2-11, 1348 Louvain La-Neuve Tel.: 3210 473687Fax: 3210 [email protected]

Dens Koen Present address unknown

Draye XavierUnité ECOP-GC Université Catholique de Louvain,

Croix du Sud 2-11, 1348 Louvain La-NeuveTel.: 32 10 472092Fax: 32 10 47 [email protected]

Elsen Annemie Lab. of Tropical Crop Improv. Catholic Univ. of Leuven, Kasteelpark Arenberg 13, 3001 Leuven Tel.: 32 1632 9603Fax: 32 1632 [email protected]

Wuyts Natalie Lab. of Tropical Crop Improv. Catholic Univ. of Leuven, Kasteelpark Arenberg 13, 3001 LeuvenTel.: 3216 329605Fax: 3216 [email protected]

Belize

Moya Carlos Banana Growers Association Big Creek, Independence Tel.: 501 523 2282 - 501 523 2611Fax: 501 523 [email protected]

Murray Alex Big Creek Independence Stann CreekTel.: 501 523 2423Fax: 501 523 [email protected]

Ramclam Wilbert Big Creek Independence Tel.: 501 523 2000/2001 - 521 2902Fax: 501 523 2112 - 523 [email protected]@[email protected]

List of participants / Lista de participantes

The problem of banana root deterioration and its impact on production254 List of participants - Lista de participantes C.A. Gauggel et al. 255List of participants - Lista de participantes

Brasil

Linz Robert Harri EPAGRI, Estacao Experimental de Itajaí, Caixa Postal 277, 88.301-970 Itajai, Santa CatarinaTel.: 47 341 5244Fax: 47 341 [email protected]

Régis Fabio Caixa Postal 227, CEP. 63011-970, Juazeiro do Norte, CearáTel.: 88 501 2201Fax: 88 501 [email protected]

Chile

Maturana Gonzalo Casilla 521, Rancagua Tel.: 5672 411 949Fax: 5672 411 [email protected]

Colombia

Belalcázar Sylvio Carrera 16 #21N-10, Armenia, Quindío Tel.: 5767 495371Fax: 5767 [email protected]

Martínez Ana M. Calle 32 C No. 88-52, Barrio Capellanic, MedellínTel.: 571 267 9464 57 - 310 263 4393Fax: 571 267 [email protected] - [email protected]

Mira John Jairo Calle 3 Sur No. 41-65 Edificio Banco de Occidente Piso 9, MedellínTel.: 5748 226 602Fax: 5748 236 [email protected]

Mora Isolina Villa del Río Cuarto Etapa #29, Apartadó, Antioquia, Zona UrabáTel.: 828 2614Fax: 828 [email protected] (cont’d)

Patiño Luis F. Calle 3 Sur No. 41-65, Edificio Banco de Occidente Piso 9, MedellínTel.: 5748 226 602Fax: 5748 236 [email protected]

Sierra Francisco Calle 16B Sur No. 42-97, Medellín Tel.: 574 313 6011 - 574 8280 036Fax: 574 8280 [email protected]

Costa Rica

Araya MarioApartado 390 7210, GuápilesTel.: 506 763 3260Fax: 506 763 [email protected]

Arias Madriz Oscar Bayer, La Uruca Tel.: 506 391 1030 - 506 243 6038Fax: 506 258 [email protected]

Bellavita Guido P.O. Box 1864-1100, San JoséTel.: 506 240 [email protected] [email protected]

Blanco Fabio CORBANA, La Rita, Guápiles Tel.: 506 262 1350 - 506 835 [email protected]

Bolaños Erick CORBANA, 28 Millas, LimónTel.: 506 718 6328Fax: 506 718 [email protected]


Costa Rica (cont’d)

Bruce PrósperoStandard Fruit de Costa Rica S.A. Tel.: 506 710 3900Fax: 506 710 [email protected]

Campos Bernant BASF de Costa Rica, Plaza Roble, C.R. Edif. Los Volcanes, San JoséTel.: 506 308 8548 - 506 201 1900Fax: 506 716 [email protected]

Campos Ricardo Standard Fruit Co., Torre Mercedes, 7 Nivel Paseo Colón, San JoséTel.: 504 764 [email protected]

Cardona Jorge Bananera El Esfuerzo, 75 mts. Sur OIJ, SiquirresTel.: 506 768 5481 - 506 718 6025Fax: 506 768 5481 506 718 6025

Chávez S. Rodrigo 100 mts norte, 50 este y 50 N Banco Popular, GuápilesTel.: 506 234 0540 - 506 711 0315 - 506 369 3126Fax: 506 234 2046 506 711 0315

Corrales Omar Ultrapark, Edificio 7B, 1 Km noreste Plaza Real Cariari, HerediaTel.: 506 293 9500Fax: 506 293 [email protected]

Corrales M. Gustavo Scotts Company, Apartado 133-7100 Paraíso de CartagoTel.: 506 350 73 52Fax: 506 [email protected]

Coto Raúl Apartado 1384-1250, Escazú Tel.: 506 228 8264Fax: 506 228 [email protected]

Duarte Ricardo Standard Fruit Co., Torre Mercedes, 7 Nivel Paseo Colón, San JoséTel.: 504 764 7000

Fallas Gustavo Santa Ana, San José Tel.: 506 204-20-02Fax: 506 [email protected]

Gámez Juan JoséBANACOL, Guápiles, LimónTel.: 506 767 7171, ext. 34Fax: 506 767 [email protected]

Garita Pablo CORBANA, Apartado 6504-1000, San JoséTel.: 506 224 4111Fax: 506 253 [email protected]

Gómez Omar Agrícola Ganadera Cariari (BANACOL), GuápilesTel.: 506 767 7171 - 506 767 8182 - 506 381 2796Fax: 506 767 [email protected]

Gómez Eduardo Apartado 11830-1000, San José Tel.: 506 296 2156Fax: 506 231 [email protected]

González Roberto Bayer, La Uruca Tel.: 506 391 1030 506 243 6038Fax: 506 258 [email protected]



González Miguel CORBANA, Apartado 390-7210, GuápilesTel.: 506 763 3257 - 506 573 3260Fax: 506 763 [email protected]

Guendel Ronald Bayer, La Uruca Tel.: 506 391 1030 - 506 243 6038Fax: 506 258 [email protected]

Guzmán Mauricio CORBANA, La Rita de PococíTel.: 506 763 3176 - 506 763 3260Fax: 506 763 [email protected]

Guzmán Arnoldo Standard Fruit de Costa Rica S.A. Tel.: 506 287 3461Fax: 506 287 [email protected]

Jara Sonia CORBANA, Apartado 390-7210, GuápilesTel.: 506 763 3260 - 506 762 3176Fax: 506 763 [email protected]

Jaramillo Ramiro Apartado 4824-1000, San José [email protected]

Lang Juan Apdo. 83-1225, Plaza Mayor Pavas, San JoséTel.: 506 220 1822Fax: 506 291 [email protected]

Madriz A. Felipe La Uruca, San JoséTel.: 506 290 0059 ext. 229Fax: 506 231 [email protected]

Marín Douglas Del Monte, Corporación de Desarrollo Agrícola, Apartado 4084-1000, San JoséTel.: 506 769 1594 - 506 769 1415 ext. 4519Fax: 506 769 [email protected]

Martínez Eduardo Colonia Sarapiquí, Junto Oficina Caribana, HerediaTel.: 506 710 2271, ext. 841 - 506 393 1936 Fax: 506 710 [email protected]

Medina S. Leslie Apartado 10929-1000, San José Tel.: 506 205 1000Fax: 506 205 [email protected]

Monees Thomas INIBAP CATIE, 7170 TurrialbaTel.: 506 556 2431Fax: 506 558 [email protected]@belgacom.net

Montero Nelson Colono Agropecuario Jiménez, Pococí Tel.: 506 763 7300Fax: 506 767 [email protected]

Montero C. Pablo 2 km Oeste del Parque Empresarial Forum sobre Carretera Próspero FernándezTel.: 506 205 1058Fax: 506 205 1010 - 506 205 [email protected]

Orlich Romano Inversiones Orlich, Apartado 2661-1000, San JoséTel.: 506 223 4363Fax: 506 221 [email protected]



Páez MarcoCristal Vitro S.A., Apartado 1084-2050, Concepción de Tres RíosTel.: 506 279 5543 - 506 279 3618Fax: 506 279 4216 506 279 [email protected]

Parra José Apartado 4595-1000, San José Tel.: 506 236 2364Fax: 506 256 [email protected]

Pérez LeonardoChiquita Forum, Edificio D, Santa Ana, San JoséTel.: 506 204 2000 - 506 385 0810Fax: 506 204 [email protected]

Pocasangre Luis INIBAP CATIE, 7170 TurrialbaTel.: 506 556 2431Fax: 506 558 [email protected]

Quesada Miguel Del Monte, Apartado 4084-1000, San JoséTel.: 506 769 1415, ext. 4302Fax: 506 769 [email protected]

Quirós Sandí Luis EARTH, Guácimo, LimónTel.: 506 713 0000 - 506 713 0361Fax: 506 713 [email protected]

Ramírez Alexander 200 N y 75 O de la Bomba Santa Clara, GuápilesTel.: 506 710 4665 - 506 355 3916Fax: 506 7104665

Ramírez Fernando Sabana Norte I.C.E., 100 metros Oeste, 150 metros al Sur (MAKHTESHIM-AGAN), San José

Tel.: 506 296-24-28Fax: 506 [email protected]

Rivas Galileo CATIE, 7170 Turrialba, Tel.: 506 558 2391Fax: 506 558 [email protected]

Riveros Alba Stella CATIE, 7170 TurrialbaTel.: 506 558 2446Fax: 506 556 [email protected]

Rodríguez Werner Apartado 2300-499, Curridabat, San JoséTel.: 506 387 3392Fax: 506 556 [email protected]

Romero Ronald Chiquita Forum, Edificio D, Santa Ana, San JoséTel.: 506 204 2000Fax: 506 204 [email protected]

Rosales Franklin E. INIBAP CATIE, 7170 TurrialbaTel.: 506 556 2431Fax: 506 558 [email protected]

Sancho Hernán Santa Ana, San José, Tel.: 506 204-20-02Fax: 506 [email protected]

Sandoval Jorge CORBANA, Dirección de Investigaciones, Apartado 390 7210, GuápilesTel.: 506 763 3260 - 506 763 3176Fax: 506 763 [email protected]



Sauma Jorge CORBANA, Apartado 6504-1000, San JoséTel.: 506 224 411Fax: 506 283 [email protected]

Segura Alvaro CORBANA, Apartado 390-7210, GuápilesTel.: 506 763 3257 - 506 763 3260Fax: 506 763 [email protected]@racsa.co.cr

Serrano Edgardo CORBANA, Apartado 390-7210, GuápilesTel.: 506 763 3257 - 506 573 3260Fax: 506 763 [email protected]

Trejos César BANACOL, Guápiles, LimónTel.: 506 767 7171, ext. 34Fax: 506 767 [email protected]

Ureña A. Rolando Apartado 10929-1000, San José, Tel.: 506 205 1000Fax: 506 205 [email protected]

Vaquero Roque EARTH, Pocora Tel.: 506 713 0000Fax: 506 713 [email protected]

Vargas Randal CORBANA, Apartado 390-7210, GuápilesTel.: 506 763 3257 - 506 763 3260Fax: 506 763 [email protected]

Vásquez Nelly CATIE, 7170 Turrialba, Tel.: 506 556 6455Fax: 558 [email protected]

Vega Lissette INIBAP CATIE, 7170 TurrialbaTel.: 506 556 2431Fax: 506 558 [email protected]

Cuba

Fernández Emilio INISAV, Calle 110 No.514/5ta B y 5ta F, Miramar, Playa, Ciudad HabanaTel.: 537 209 3683Fax: 537 202 [email protected]

Ecuador

Carrillo Rodrigo Av. Francisco de Orellana, Edificio Centrum, Piso 11 Ofic. #3, GuayaquilTel.: 5934 296 0151 - 5934 268 1799Fax: 5934 296 1051 5934 268 1899

Cevallos Mario Av. Francisco de Orellana, Edificio Centrum, Piso 11 Ofic. #3, GuayaquilTel.: 5934 268 1799Fax: 5934 268 1899

García Xavier Av. Francisco de Orellana, Edificio Centrum, Piso 11 Ofic. #3, Guayaquil Tel.: 5934 268 1799Fax: 5934 268 1899

Palomeque JuanBolívar 712, Machala, El OroTel.: 593 99 746814Fax: 593 72 934 [email protected]


Ecuador (cont’d)

Sandoval GuidoKm 9 1/2 vía Daule, Guayaquil Tel.: 5934 211 0500Fax: 5934 211 [email protected]

France

Staver Charles INIBAP, Parc Scientifique Agropolis 2,34397 Montpellier Cedex 5 Tel.: 33 467 61 13 02Fax: 33 467 61 03 [email protected]

Germany

Sikora Richard University of Bonn, Soil Pathol. and Nematology, Institute for Plant Diseases, Nussallee 9, D-53115 BonnTel.: 49 228 732 439Fax: 49 228 723 [email protected]

Honduras

Castro MarcoStandard Fruit Honduras, Departamento Investigaciones, Colonia El Naranjal, La CeibaTel.: 504 443 [email protected]

Gauggel Carlos Escuela Agrícola Panamericana, P.O. Box 93, Zamorano Fco. Morazán, TegucigalpaTel.: 504 776 [email protected]

Lobo José Standard Fruit Honduras, Departamento Investigaciones, Apartado 96, La CeibaTel.: 504 442 2787Fax: 504 442 [email protected]

Pocasangre Hugo Standard Fruit Honduras, Departamento Investigaciones, Apartado 96, La Ceiba Tel.: 504 442 2787Fax: 504 442 [email protected]

Portillo CarlosStandard Fruit Honduras Departamento Investigaciones, Colonia El Naranjal, La Ceiba Tel.: 504 442 [email protected]

Israel

Israeli Yair Jordan Valley Banana Exp. Station, Zemach 15132Tel.: 972 4 675 7670Fax: 972 4 675 [email protected]

Jamaica

Garwood Elaine c/o Banana Research Dept. 10 South Ave., Kingston GardensTel.: 876 922 2083 - 876 967 5592Fax: 876 967 [email protected]

Mexico

García Erasmo A. Palmahuaca Produce Km. 2.5 Carrt. Teapa-Sn Antonio, Teapa, TabascoTel.: 932 32/20475/20323Fax: 93232/[email protected]

Puerto Rico

Díaz Manuel Estación Experimental Agueda, P.O. Box 1306, Gurabo, Puerto Rico 00778Tel.: 787 737 3511 - 787 712 0470Fax: 787 737 [email protected]

The problem of banana root deterioration and its impact on production260 List of participants - Lista de participantes

Spain

Jaizme Vega María del Carmen Dpto. Protección Vegetal, Apdo 60, 38200 La Laguna, Tenerife, Canary IslandsTel.: 34 922 476356Fax: 34 922 [email protected]

St. LuciaLloyd Davidson WIBDECO P.O. Box 115 CastriesTel.: 758 452 2411Fax: 758 451 [email protected][email protected]

Uganda

Blomme Guy INIBAP P.O. Box 24384, KampalaTel.: 2567 748 3077Fax: 2564 128 [email protected]

USA

Caid Russel D. 988 Chambers Road, Walton, Ky 41094-9522 Fax: 1 859 485 [email protected]

INIBAP – Red Internacional para el Mejoramiento del Banano y el Plátano La Misión de INIBAP es aumentar de manera sostenible la productividad del banano y el plátano cultivados por pequeños productores para el consumo doméstico y mercados locales y de exportación. El programa tiene cuatro objetivos principales:• Organizar y coordinar un esfuerzo global de investigación sobre banano y plátano para el desarrollo,

la evaluación y la diseminación de cultivares mejorados y para la conservación y utilización de la diversidad de las Musaceas;

• Promover y fortalecer colaboraciones en la investigación relacionada con banano y plátano a los niveles nacional, regional e internacional;

• Fortalecer la capacidad de los SNIA para conducir actividades de investigación y desarrollo sobre banano y plátano;

• Coordinar, facilitar y apoyar la producción, recopilación y el intercambio de información y de documentación sobre banano y plátano.

INIBAP es una red del Instituto Internacional de Recursos Fitogenéticos (IPGRI), un centro ‘Future Harvest’.

MUSALAC – Red de Investigación y Desarrollo de Plátano y Banano para América Latina y el CaribeMUSALAC surgió mediante un Acuerdo de Constitución en el ámbito de FORAGRO, firmado el día 6 de junio del 2000, en Cartagena de Indias, Colombia. Los socios de MUSALAC son 15 instituciones nacionales de investigación y desarrollo en representación de sus respectivos países (Bolivia, Brasil, Colombia, Costa Rica, Cuba, Ecuador, Honduras, Jamaica, México, Nicaragua, Panamá, Perú, Puerto Rico, República Dominicana y Venezuela) y 4 instituciones regionales/internaciones (CATIE, CIRAD, IICA e INIBAP). El objetivo general de MUSALAC es incrementar la productividad y competitividad en la cadena agroalimentaria del plátano y el banano a través del desarrollo científico y tecnológico, fortaleciendo los sistemas nacionales de investigación y desarrollo, integrando actores, priorizando y coordinando acciones en América Latina y el Caribe. MUSALAC está dirigida por un Comité Directivo conformado por un representante de cada país miembro; un Presidente y dos Vice-Presidentes y una Coordinación Ejecutiva, actualmente coordinada por INIBAP-LAC, con sede en el CATIE, Turrialba, Costa Rica.

CORBANA – Corporación Bananera NacionalCORBANA S.A. es una institución pública sin fines de lucro, cuyos objetivos son:• Fortalecer la investigación en el cultivo del banano.• Incrementar la productividad bananera con un mínimo de riesgo ambiental. • Fomentar programas de reducción de costos. • Brindar servicios de investigación, asistencia técnica e información sobre precios y mercados. • Propiciar un régimen equitativo de relaciones entre productores nacionales y empresas

comercializadoras.• Establecer junto con el Gobierno, políticas bananeras que ayuden al mantenimiento de la industria

a largo plazo. • Centralizar la información sobre banano para promover y fomentar la participación en la investigación

y en el desarrollo tecnológico del sector bananero.

INIBAPParc Scientifique Agropolis II34397 Montpellier Cedex 5France

MUSALACC/o CATIEApartado 607170 Turrialba, Costa Rica

CORBANA S.A.A.A. 6504-100 San JoséCosta Rica

Banana root system: towards a better understandingfor its ...

Documents

Transcript of Banana root system: towards a better understandingfor its ...